#💡┃suggestions-web

1 messages · Page 26 of 1

void shadow
#

Since the algorithm can differentiate between different stems, is it possible to add a feature that can change certain aspects of the song but keep others? Like change vocals but keep the instrumental or vice versa?

opal spoke
#

Phone ringtone genre

glass barn
weary birch
#

I think this isn't a new suggestions but I wish there was more option around the trash and delete topics. Like selection of everything listed instead of either clicking everything or command+a and maybe the option to delete everything in the trash after x days or auto-delete things older than x days. It's really a pain in the eye seeing a list of 30+ consisting of minor parts for the final song you finished and than deciding to clean up because of your OCD 😦

glass barn
weary birch
glass barn
#

Cut up a few bits that you like make it go six seconds long or start with 4 second silence then 2 sec sample

#

Ooooor Just learn how to use fm studio

weary birch
# glass barn Ooooor Just learn how to use fm studio

FM studio? I really want to keep it a simple (yet sometimes very time consuming) hobby without any more programs right now, but thanks for the tips. I still wish that suno could implement a simple "select all" feature 😁

void shadow
#

also I have a BIG suggestion please. I made a song that is like 6 parts and kind of intricate at like part 3 I messed up and extended the wrong clip. (sometimes the part finishes and moves on to the next and then I accidentally like and extend that part instead of the intended part) So messing up in Part 3 but parts 1, 2,4,5,6 are fine. I tried to use the crop and all the tools available to me to fix what I can but it's not to my satisfaction at all which basically means I have to remake the entire song (I used about 5000 credits perfecting this song so I'm quite upset :/)

My Idea is pretty much the opposite of the current crop feature.
It would function mostly the same but instead of cropping what you want to keep you could crop what you want to change. I believe they call that inpainting in AI art generators.

I really hope you consider this as it would be a MAJOR help.
I'm so upset with myself that I let this happen 😦

silver nacelle
#

@tough walrus how can I upload just a voice e sample without it turning the audio clip into a song?

glass echo
#

A star system would be cool. That way when you are trying to figure out which part or song you like the most you can mark them as 5 stars down to 1 star. The thumbs up and down doesn't help for those songs/parts that you need to do a A|B listening test.

#

Another suggestion would to lazy load the past songs listed in the Create section. This would be helpful when using an iPad where loading a new tab is a hassle just to reference/listen a previously created song/part.

#

Also, why doesn't the Library show the parts highlighted? I find my self searching for the other parts when having more than one or two. Especially when to the A|B listening test to find the perfect variation.

tough walrus
civic cedar
#

When autoplaying, the selection should automatically move to the next song that is playing, having its image and lyrics in the right side panel, instead of being stuck on the first song you selected.

hollow tapir
#

can you make a button that makes it that tempo changes are not allowed?

i made a song i wanted a hammer beat under, but when eddeting it, it had 9 tempo changes.

120 bpm (my start song to extend) all the way to 136 bpm, realy anoying

languid sky
#

anyway to add a "redo" or "recycle" for one of the tracks that populates? seems like there is always 1 out the 2 that is just so off its nearly an auto delete within the first 10-20seconds on an extend

#

however in the same sense, the one that is very good a lot of times comes with areas that are really scattered and all over the place. if we can even choose the section of the song we would like to "repopulate" that would be great!

remote pond
#

It would be nice to suggest a melody to Suno, and have it build an arrangement around it. This would allow me to then suggest a style for it.

For example, I'd like to hand Suno the melody for Good King Wenceslas and have it turned into a marching band arrangement, for use in the background of a Christmas Parade video.

languid sky
#

seems like the more i extend a song it gets more and more distorted and vocals get very wavey like their is a phaser effect on the track

tough walrus
languid sky
#

when extending a song and you choose instrumental only, it still throws in mix phrases from the original track's vocals. can we get it to where it really is just an instrumental extension?

velvet horizon
#

"This song violates Suno's content moderation policy." You know for once I wish that a system like Suno would just tell me what the violating part is because I absolutely have no idea. If I knew what the "offending" part was I could fix it.

tough walrus
robust jewel
#

Suno can't tell you what part is the problem because ChatGPT doesn't actually tell them, from what has been said in the past. It just refuses to deal with what it was given. You should be able to ask ChatGPT if there are any aspects of the lyrics that it can see. Typically it will either see something as violence (especially toward children), overly sexual, or race related.

#

And ChatGPT can be a little overly sensitive at times.

modest frigate
#

I understand that the model writes all instruments and the voice in the same time. So there are no separate tracks, not even in the memory I guess. But I have seen from the Cover function also that the model can copy any instrument, all instruments and even the voice melody from a song. This means that separate voice generation could be already eventually achieved this way: to "force" the AI with the help of some commands to post generate, for a certain song, the same voice (or a similar one) in the same way (pace, timing, bpm etc). To be able to do this the model has to mark in the song the exact positions of the different parts (verse, chorus, the bpm, the key etc), the notes and the intonation elements and to compare at the end with the original. But I guess the model must learn this process of copying the voice or an instrument first (training). And I guess it can do it easily.

#

ChatGpt is discriminating 🙂 I made some funny songs about different nationalities, which was not a problem. But about japanese, Chatgpt did not allow this.

#

Would be nice and professional to have something similar from Suno too: Udio's feature requested from users.

image.png
modest frigate
lyric acorn
# void shadow also I have a BIG suggestion please. I made a song that is like 6 parts and kind...

1000%. Having a bunch of cropped files is nice but if you can't combine them back into a song then it's relatively useless. Any cropped file (at least from the same style prompt) should be able to extend from each other in whatever order you want, with duplicates if desired. Allowing the Suno AI to join them as best it can (can use credits to join with transitions, I don't care). I have a 7 part song that, due to a bug, the associations were lost and so Get Whole Song just results in a mess. Allow me to tell Suno which is part 1, part 2, part 3 etc. If I don't like part 4 I'll make part 5 the new part 4 etc. Maybe use the intro part again after that if I like it, before continuing to part 6, then I'll duplicate part 3 (which had a particularly good chorus section that Suno doesn't want to generate again) as part 7, you get the idea

#

Pretty sure Suno could find a way to join death metal with twee-pop in a seamless transition, so the style prompt limitation would probably be temporary if it's there at all

frigid pivot
#

**Feature idea. **

Ability to connect any two songs in your library, using something like "Extend to". How this could work:

  • Choose "Extend to" from context menu
  • Window opens up with an option to set a timestamp at which you want to extend from, then pick the other song from a searchable menu
  • Once the second song has been selected, choose an entry point timestamp (where the first song extends to)
  • Then adjust a slider at the bottom of the window to determine how many seconds the AI has to work with to manipulate and mesh the songs together, like a range of 5 to 20 seconds.

Obviously it would work better most of the time with 20 seconds, but if the songs are similar enough or the styles are typically seen together, shorter could work. The user would have to play with it a little to see what works for each song, as we do already for the other features.

leaden marlin
#

Song report channel

velvet horizon
#

"This set of lyrics doesn't seem to contain content that would violate ChatGPT's guidelines. "

lavish sleet
steady river
#

I'd love to be able to import an instrumental track of my own up to ~3 minutes and rather than extend it, keep the track and add vocals.

lavish sleet
#

Feature Request: Separate Album Option

Currently, Suno.ai offers a playlist feature that allows users to group their songs together and correlate them with others. However, it would be helpful to have a separate Album option, specifically designed for releasing collections that feature one artist. This would allow users to create and release albums in a more structured format, distinct from playlists, making it easier for artists to organize and distribute their music as complete bodies of work.

tidal wharf
#

guitar hero stems to the rescue to train a model 🙂 (not sure if there's enough of them though)

steady river
#

It would be fantastic to be able to generate a vocal stem together with instrument specific midi files ....

naive solstice
#

Some AI platforms, already have AI-generated track shops. They allow non-creators to buy tracks, which increases website traffic and sparks interest among creators.
Do you planed do this too?

#

And what about INPAINTING selected part of song?

shrewd pike
#

}

#

saving and reusing a vocalist would be amazing!!

#

Also I really want to be able to create an embeddale radio station from my creations.

hollow tapir
#

do not let suno let a verse use multiple times, im getting realy anoyed, at many songs that have 3 verses, but verse 2 gets repeated and there is no time for verse 3

tepid trench
#

Might be nice to scale the threshold for automatic credit refunds, based on model version.

Current threshold (unchanged since v2 I think?) seems to be around 8 seconds which is roughly 10% of the max length of v2 first part.

So, based on this I'd suggest just having 10% of the max first part length as the threshold:
10% of 2 minutes (v3) = 12 seconds
10% of 4 minutes (v3.5) = 24 seconds

Getting short generations that are slightly above the existing threshold for models that are intended to give longer generations can be frustrating, particularly when a set of prompts is continously generating short songs (some of which are below the threshold, whilst others are not.)

hard tinsel
modest frigate
#

THE MOST ANNOYING UI PROBLEMS NOW

  • hard to detect playing song due to very bad visibility of that
    animation. The song playing needs to be clearly highlighted.
  • the like button in the search mask is not working properly
  • creating songs with the same name (add a counter)

THE MOST ANNOYING AI PROBLEMS NOW

  • cover delivers songs you don't want actually or the same song over and over again. Give us parameters to be able to control the function.
  • not being able to finish a song even when you use [End]
  • AI model not respecting the lyrics and the song structure
  • we need a "FIX" function to only fix the sound with a new identical generation and this should work also for whole 4 mins songs.
#

THE MOST NEEDED MISSING FEATURES

  • possibility to hide the "parts" songs
    Folders: Just ADDING folders would be a MISTAKE! Cause the creation process has to be changed, it has to happen in a folder so you don't need to move songs after creation. It has to be similar to a PC. You don't put there your all files in a single folder. You make a folder for a certain song/genre and all generations go there
    You make e.g. a "salsa" folder and you create only there salsa songs. Or you make a folder called "My Song Loving you" and you make there all the generations for this song
    When a song is finished you move it into the folder "Finished songs" or "Rock songs" or whatever
    This way you have the best organisation. The folders should be visible in a separate column window (best would be left side) so you can navigate easily (like in spotify)
    We need drag and drop function to move songs from right to left (left button mouse will lift so you can put in one folder) or some mass moving function where you select more songs at once
  • marking good songs 1-3 or 1-5 stars
  • buttons in the song line for: download, changing title, get whole song (goes faster)
  • possibility to add a note to a song for a later review
  • possibility to download many songs at a time (wav, mp3, vid)
  • possibility to reuse the last prompts for style and lyrics again without typing them again (Udio has some)
  • a normal refined search function (hide "parts", look only in the title, AND+OR operators for the keywords.
  • possibility to mark and search for songs with a "needs rework" symbol for later finding/editing
  • tree structure display for the "parts" (icon offset to the right)
  • other color for the "full song" button
  • shuffle function for the player
  • go to next/previous song via keyboard
  • show style/genre the interpreter sent to the model (like Udio does) after gen.
  • we need a "liked" filter on the create page
#

THE MOST NEEDED AUDIO FUNCTIONS

  • Inpainting - this can be used to change words, lines, verses or to sing different a certain passage
  • parameters for the cover function that give us some control over this
  • fading out a song
  • extending a song on the "left side", like Udio does. This is needed when you need an intro that was missing or was not good but you dont want to change the rest of the song 1
  • cropping the uploaded songs automatically to maximum length when these are longer.
  • importing 4-5 minute songs for real covers
  • a parameter for the voice volume (which is sometimes too high)
  • MERGING VERSIONS OF THE SAME SONG: copy from other generation of the same song to place in the one you work at. Cause there are sometimes better choruses or intros or outros in other generations that won't repeat anymore and you will never get otherwise. MERGE: you have the waveform of the destination song and you chose the insert point, you open another song (the source) with the waveform and mark the loop that you want to copy it /merge it in the destination.
#

NICE TO HAVE

  • show the BPM of the song in the song icon
  • direct access to the AI model and seed control
  • some parameter to control the interpreter
    which feeds the AI model
  • displaying how much from a song you
    listened in a graphical waveform as in Splice (this way you can spot unlistened songs)
  • a play button and icon in the left down corner for the files you have chosen to cover so
    you can do A-B compare (like in extend mode)
  • floating and scalable window for the lyrics so you don't have to deal with the tiny one like it is now
  • fast audition which play the songs automativally not fully, but only a number of seconds you set in the pref.
  • a random style out of many from a huge list (made by suno), it helps make unusual music
  • user preferences page to save certain settings
  • show the musical key of the song
  • to be able to pick up the same voice from a certain song made already
  • to be able to look for a song in playlists
  • a list with all uploaded songs
clever harness
#

Additional Meta Data

I'd love to know time to like (i.e. the delta between clicking play and clicking the like button) in the meta data. In reviewing some of my data, I can see some people like tracks faster than is possible for the human mind to process the sound waves. I think I'm good, but I don't think that I'm thaat good, so I'd like to filter possible bot activity.

amber finch
#

been a while since I tried out suno. loving the cover feature.

new features to consider:

  • provide vocal sample / seperate prompt for vocal style/seed control. having vocals in songs is nice but it is next to impossible to get the same one twice even if you provide your own samples as a start with the current feature
  • key/scale control. as it suggests, be able to suggest a key or scale for the song to be restricted to. (or maybe even multiple keys/scales for those that key-change)
  • song seed control. if suno works like other diffuser ais I assume that each song is generated with a seed. being able to change the style and lyrics but keep the seed offers potentially more control over the generations

EDIT:

  • also since covering is a thing now, denoise control (assuming the cover system uses denoise for generation)
civic cedar
#

Make music style and vocal style separate settings.

sharp meadow
#

  1. A nice thing to have would be to be able to trigger a point where the lyrics start. Right, what if I want to make a shoegaze, or an orchestral type song, but not have the lyrics start immediately? What if I want it to be an instrumental for like 7 minutes, but have the lyrics start at 8 minutes?

  2. let us modify where we want a climactic point in the song to be. Let us like....modify where each section of the song starts and ends, the little in betweens before each section starts. Custom section names and labels. Let us have that amount of customizability to our songs.

  3. Let us prompt certain instruments for different sections of a song.

onto vocals, which @civic cedar above me mentioned, which I'd like to expand on.

  1. Allow us to customize the AI vocals. Let us select specific vocal styles, tones, etc.

  2. MAybe even let us layer vocal tracks. Like, create a new section on the website designed for creating AI vocals, let us take those vocals, layer tghem in the site, and allow us to make harmonies, backing vocals, or AI duets.

  3. Let us add vocal effects like reverb, autotune, pitch correction, and distortion on to the vocals. Some people don't have the money to pay for a DAW to pull the stems and manually do it.

  4. Make ready-made templates based on various genres for new users to get used to the system easier. Sure, we have a way to explain to the AI on what we want right in the creation page, but make templates that new users can pick and choose, modify.

  5. Have the AI give suggestions for song arrangement especially regarding lyrics, based on the genre and styles that the user selected, allowing users to optimize their songs flow and structure.

  6. Make a tool, like an AI, that suggests key changes, time signature adjustments, and tempo changes to smooth transitions between sections of the song. Add new tools to actually add and modify those.

  7. To REALLY make Suno a viable place to make music, have the AI function as a "co-creator", and have it collaborate with the user by providing feedback alongside the user.

  8. Provide in-site mastering tools.

  9. Lyric improvement suggestions.

  10. Allow the user to train the AI, and guide it to create more songs in that specific style.

  11. Finally, AI critique. After a generation, the AI could give feedback on aspects such as rhythm tightness, depth of the lyrics, melodic complexity, and suggest improvements. This all ties in with the other 13 changes I suggested.

broken lagoon
#

I was thinking that a good QoL option would be to be able to filter songs in the library by whether they're an uploaded clip rather than remembering the specific name or hunting through pages and pages of songs

floral sandal
#

Suno really, really doesn't want to give me guitar solos lately.

floral sandal
#

and why are my cover songs of a 3:49 song coming out to 3:20 man this feature was not even ready for beta

modest frigate
#

3:20-3:25 seem to be maximum what the cover function can make

shell valley
#

When i download MP3, i miss the picture.

rocky spire
#

I would like to be able to recreate the same song after a lyrics treat or an arrangement change. Do I just not know hot to do that yet?

floral sandal
#

The cover feature was supposed to be able to do that but I haven't succeeded. I wanted to change just a single word in one of my songs to make better sense with the progression but alas.

sick lion
#

Civitai (AI art generator) has a "Buzz" system where you obtain a certain amount per day just by liking others posts (there's more to it but that's the important bit). If Suno did a thing based off LISTENING to other people's songs it could 1) get others exposure and 2) potentially free up time (resources) because people are listening and not generating (maybe?).

#

Putting this here too:
It's definitely a balancing act for AI generation tho - from what I hear it's intensive, at least for art. Maybe there could be radio stations that people submit their posts to that expire (remove) after a time (ive seen "go to song radio" on the post but that's more an automated thing based off tags I think). Could be like a forum or group sort function. Or maybe people could create their own stations idk losing my train if thought.

magic granite
#

If you guys somehow configured a system that allowed for sampling, be it for specific instruments, sound samples, etc? I think I'd lose my mind. NEEEEEEEEEEED.

sick lion
#

Allow links in playlist description (to link covers to original album, or a sequence of playlists that go together).

steady tinsel
#

probably already mentioned but if we could get an option in extend to pick a range of extension, that would be great! when you need to tack on an extra 20-30s (or whatever) and its generating 2+ min clips whne you have 3-4min already is annoying.

marble vessel
#

In playlists, can you add - Download all selected, Move all selected to another playlist options so that we can do bulk ops on a playlist?

ionic nova
#

I've already written about this here and in direct emails to support. So far no one understands what's going on.
This is my link to my post.
#🚧┃suggestions-layout message

steady tinsel
# ionic nova blob:https://discord.com/98fd227d-c591-4e08-8c43-040d8c7e4fe3

for the love of god yes, some kind of organization. I have way too much stuff from how I pattern my generations and extensions so the list just becomes so daunting that its hard now to keep track of things. if you jump and work on somtehing else its easy to forget about something because its 5 pages down because you had a "fun night" and went on a rampage lol

modest frigate
#

IMPLEMENTING FOLDERS: if Suno implement folders it should not do it superficially! Just ADDING folders would be a big MISTAKE!
Cause the whole creation process has to be changed, it has to happen in a chosen folder already so you don't have move around songs and parts after creation.
It has to be similar to the folders of a PC. You don't put your all files in a single folder there like is it now in Suno.
You make a folder for a certain song or genre and all generations go and stay there. You work in a certain folder every time.
You make for example a "salsa" folder and you create only there salsa songs.
Or you make a folder named "creation folder" where you make your generations (this should be the default folder).
Or you make a folder called "My Song Loving you" and you make there all the generations for this song.
When a song is finished you move it into the folder "Finished songs" or "Rock songs" or whatever.
This way you have the best organisation. The folders should be visible in a separate column window (best would be left side) so you can navigate easily (like in spotify).
We need drag and drop function to move songs from right to left (left button mouse will lift so you can put in one folder) or some mass moving function where you select more songs at once to move them for organisation.
There must be also a filter to filter out "parts" songs for a better visibility (show parts, hide parts).

modest frigate
#

The later Suno implement folders, the more work we all have to do to reorganize our library! I just wonder how Suno and Udio forgot to think about song organisation
from the start and gave us only a single place to put thousand a songs without any means of organisation. Any mp3 player has folders and and any audio software or site.

near fern
#

I really love the cover feature, but it really needs a control strengh slider. Sometimes the song is not changing at all, sometimes it's very different

#

often I have to remove 80% of the generations cause they are identical with the covered song

patent nacelle
#

Feature request : remake the exact same song with different lyrics.
The cover feature could potentially work for that but it doesn't seem to change the lyrics at the beginning of the song, and the output is still different from the original. What would be cool is if we could simply input new lyrics in any given language (so we can make adaptations in a different language) and click "Re-generate the song", it will then create the exact same one with the new lyrics.

Useful for :

  • Correcting lyrics on an existing song (mainly because the song is good, but we want to change part of the lyrics to make it sound better)
  • Making translations or adaptation in a foreign language (like a Disney song lol)
floral sandal
river prairie
#

Hi all

#

I was trying to get stems for a song

#

But the stems I get appear to be frequency-based, which means the instrumental one has a lot of voice "echo" and the voice one is pretty poor quality

#

Is there a way to get "real" stems?

#

If not, would be a real nice feature to have, as personally I find the current stems rather questionable for any practical use ...

hybrid crane
keen hollow
#

Been using Suno for a really long time and it's helped my actual music creation process. However, since I generate so many variants, I like cleaning up my library.

Biggest weakness at the moment is that I can't mass delete anything in my library.

Can we get a way to select multiple in a search or some way to delete massive amounts of songs at once?

river prairie
tepid trench
keen hollow
civic cedar
#

Show the secret, enhanced prompt for user education.
Some AI Art apps show you the "enhanced prompt" randomly selected for you by the app.
Suno is clearly doing prompt ehancement as well, for the same reason, to get a good variety of results.
But Suno hides what the random secret extras were, so we can't learn from the AI as we go.

tidal zinc
#

This channel is already hard to follow, so I can't read everything, so I'm going to suggest something I've suggested before and I believe I'm not the only one.

• I would like to have the "move" option, e.g. move to a playlist or folder, that way the library is cleaner and separates my completed songs

dull patrol
#

Multiple Downloads

marble iris
#

For me, it would be good if you made samples for songs, there are times when I have the lyrics (which was once, because I only made 2 songs) but I don't have a good sample to put in.

echo vortex
#

I've suggest it, it's been suggested before, please allow us to rate our songs, 1-10 would be great so we can organize by rating.

timber cove
# echo vortex I've suggest it, it's been suggested before, please allow us to rate our songs, ...

If you are looking to rank to organize what you are working on... I have been making a private playlist for the song I am working on. Then I name each one with a number like "Song Name (#1)". Then in the playlist, you can order the songs to rank them against each other. The numbers help distinguish between the different variants, playlist keeps it in one place, and then the order provides the ranking beyond 1-10. Hope this helps 😊

(I only add the ones to the playlist I think are even worth it. I have noticed that if you add the song too soon, it will not actually add to the playlist. But, if you wait about 5s after the option appears it works every time.)

fallen crater
#

I suggest that there can be a way to change lyrics in songs you already made ... have been trying to do this with cover songs, but it simes like the IA mostly concentrate about the music etc. But being able to change the lyrics a bit after you made the really good song could really be a game changer to making perfect songs. Because most of the times when you make songs there will be some changes a long the way until the text is perfect and it is really bad if you have to start from start again if the start dident quite work as you wanted and everything ells is perfect.

languid sky
#

If we permanently delete tracks that are basically unusable because AI is not perfect... can we get at least partial credits back? so if 1 track costs 5 credits maybe we get back 1 or 2 credits since it can take 20 runs to get a good track 😄 ps love suno

echo vortex
# timber cove If you are looking to rank to organize what you are working on... I have been ma...

That's a good idea to help while you are working on songs. My problem is I have thousands of songs now, some of which I like, some which I really like, some of which I want to work on in the future, some of which if I tweak it I will like it, etc. So if I give everything a ranking of 5 which are songs I want to work on, and lets say 6 for those that just need tweaking, 7 are ok songs, 8 are good, 9 are bangers, 10 is perfection, etc... Then I can just organize by rating and find all my 10's, or all my 5's, anything below a 5 is not worth working on, but might have some worth as a sample, rating 1 are pure garbage. It just gives me power to organize, workflow changes, quality of life improvement big time. Keep in mind the bigger your playlist grows the more important this feature becomes.

echo vortex
river prairie
#

Feature request (I guess?): Would it be possible to ask the engine to generate 24bit, 96KHz versions of the WAV downloads, possibly for "extra credits"? Asking as it would be great to have "studio quality" masters for editing headroom.

#

(What I get right now is 16bit / 48Khz, which is close to standard "redbook [CD] quality")

leaden marlin
#

Mix your songs into 2 so basically your mixing them together

frosty breach
#

Please make it so that I'm able to select and delete both part and full songs. It's rather frustrating to have to deselect all full songs when doing a clean-up.

#

It's unnecessary how hard it is to remove full songs

modest frigate
# river prairie I was trying to get stems for a song

try lalai.ai, is the best.. You can split for free, but you cannot download, though you can record the audio and save the file with a audio downloader browser plugin, If you cannot afford paying. But even lalal.ai cannot separate perfectly the stems from Suno because of the poor audio quality and background noise.

naive moon
#

a chat box for users to chat to each others if that is possible?

tidal zinc
waxen axle
#

I dont know if this is the correct place. Time ago it existed a channel for language suggestions and the website redirects to the discord but to none channel. (If there is another place please point me).

Suggestion:
Add Galician Language

Some data about it:
Galician (Galego in Galician) is a Romance language of the Galician-Portuguese subgroup spoken mainly in the autonomous community of Galicia, Spain. It is closely related to Portuguese, with which it formed a linguistic unit (Galician-Portuguese) during the Middle Ages.

Around 3.2 million speakers.

https://en.wikipedia.org/wiki/Galician_language

Galician language

Galician ( gə-LEE-shən, gə-LISS-ee-ən; endonym: galego), also known as Galego, is a Western Ibero-Romance language. Around 2.4 million people have at least some degree of competence in the language, mainly in Galicia, an autonomous community located in northwestern Spain, where it has official status along with Spanish. The language is also spo...

deep bramble
thorn quarry
#

I dont know if this is a good section for this, but it would be nice to have delete all permanently in trash bin. Not just going one by one...

glossy ore
#

Suggestion for kind of backend-ish feature - (Accurate) BPM detection. I kind of figured that this wasn't implemented on the back-end since it's nowhere to be seen on the front end.

I've got one song that I've tried using BPM detection from different websites; none of them are accurate, nor do they agree with one another.

woven valley
hallow crown
#

I wish you could hear a voice and legit keep using that male or female singer in all your future song generations.. that would be a nice feature.

timber cove
fair bone
#

I think they should have free credit for this app because it honest cool and fun😁👍🏽 i can't have a credit card do to disability but it be nice to have free credit🙂 oh what about free people get 100 and if u want more u pay🙂. An for free people it shows a timer for how long recharge take for remind

sturdy bluff
#

How can I make a piece of music with SUNO of 15-20 seconds maximum?

modest frigate
#

with a sudden finish or fading out?

arctic cave
#

Not sure if this has been mentioned before, but it would be a huge quality-of-life change if there was something to tell you which playlists each song is in. I find it time consuming to skim every playlist to see which song is where versus if it told me in the "Add to Playlist" menu which ones it's in.

It would also be really cool if you could add a "Comments" section for each song, and give creators the ability to delete comments (such as hate comments and whatnot). However, this would also probably need something along the lines of a language filter to help auto-moderate these comment sections (to help with preventing excessive hate, spam, or to prevent people from breaking Suno's ToS within these comment sections). Doing this would strengthen the community of Suno by offering a place where not only can creators get feedback on their songs, what's working and what's not, but also providing a place to leave warm comments and reassurances for new artists. Also, if you required an account creation to comment on songs, it might also boost your user count in the long-run.

#

One last thing; It would be amazing if you could add a "Notifications" section where new likes for songs and also follows are visible. It would be really cool to see which songs are getting traction and the new peeps who are following you.

arctic cave
# fair bone I think they should have free credit for this app because it honest cool and fun...

I believe you get about 50 free credits a day for free, which is good for 10 free songs (5 generations) if you sign-up for an account. (Not sure on the mobile app, but on the mobile browser and pc browser it should give you 50 a day.)

It gets very limiting when you're working on larger projects or if a song won't come out quite-right (I spent a few days each on some of my older works because of the limited credits) but it will get the job done. I definitely recommend upgrading if you'd like to take it to the next-level, but if you're just playing around, keep creating and each day you'll have some more!

delicate cedar
#

Just remember...never make bangers on a free plan.

north iris
#

I think we need an auto-tagger for the music. If you are familiar with stable-diffusion, you may have heard of deepbooru or clip, that can reverse-engineer prompts. Every time I hear a beautiful melody, I always want to create a music piece with a similar style. The problem is that, as a music novice, I don't know what styles, techniques, or even instruments are included in the music I hear, and I’m unsure which prompts correspond to that music. It will help a lot if suno or the community can provide a music tagger. PandaLove

fair bone
raw mist
#

Please add a feature that ensures the vocals are crystal clear and undistorted.

Thank you.

sturdy bluff
tight heart
#

Is there a way to organize and even delete all of the deleted songs? Right now you can only delete 1 at a time. I have over 700 deleted songs.

clever harness
past tiger
# tight heart Is there a way to organize and even delete all of the deleted songs? Right now ...

Right Now, this site needs folders, NOT JUST playlists. If for whatever reason it's intentional to not allow multiple deletes while in the library (as a precaution rather than being intentional overall), I think at least adding a folders feature to keep songs from cluttering could help tremendously - while also simultaneously giving the option to delete the folder with all the songs in it.

rustic radish
#

Remove vocals and export. So obvious and simple to implement. Sometimes the vocals are close but not there and a human should be able to use the generated track to sing over. Simple, obvious, and again very simple to implement. Please. Thank you.

near drum
#

has the site performance on desktop taken an absolute nose-dive for anyone else lately? the site lags just trying to scroll in my own profile

topaz ruin
#

Please add a feature to export the drums as a MIDI file in the GM layout. This would allow users to export drum MIDI files, import them in to their DAW of choice, and go from there to personalize.

ruby owl
#

i would love a slider that gives the AI either more creative freedom in making a cover or making it stick to the original track more

deft pecan
#

Please Please add a option to empty trashed songs and extensions cause there are times I forget that about the trash and it get frustrating deleting them one by one.

kindred hull
#

One little fix. When the user creates a song and hits Play (from the Create view), ONLY the songs created until that moment will play. That is, all the songs created after hitting the Play button will not be played. Expected result: if you click Play from the Create view, the player should play all created songs, until there is no new songs created.

royal venture
#

Let us prompt an audio upload or cropped piece of audio within a song anywhere in the song, using a meta tag like so = <Case Sensitive Song Title>. This would allow us to make samples from our generations, or upload / reuse sound effects we can make outside of suno. Even better if the tag can be followed up by a specific time stamp like [01:45} so Suno and the creator both know exactly when the audio should be played.

This would benefit anyone greatly who wants to make atmospheric music, as well as those who want to add embelishments to existing songs (via a cover or extension) or reuse something they liked from a previous generation, but outside of the context it was generated in.

thick finch
#

Teach Suno AI some music theory for real

#

Song styles are NOT enough

Some people need specific stuff like 6/8 time, Key change, Specific chord progressions, etc...
And Suno AI just does not understand any of those

rustic radish
open pine
#

I'd love the ability to remix tracks. Something inbetween extend and cover where you can change the sound of a song without completely changing it.

raw mist
#

Its so discouraging to put the time, effort and credits into getting a really good performance but its unusable because the vocals in Sudo are distorted.

**Please make your vocals clearer. ** Your competition has done it. Sudo generates better music and performances than the competition. Make this a priority please.

Thank you.

cosmic lodge
#

Could we have extend for before a song part?
Other platforms have this capability

whole girder
#

Please add a feature to rename a multiple songs at once. Like when you select multiple songs to delete/add to playlist

ancient dome
#

A good idea for a resource, even if it is private and only for Pro and Premium users.

"Recreate" (in quotes) a song that we love but with different lyrics. For example: It doesn't have to be the same voice and the same musical instruments, but it has to be the same idea of ​​musical structure, and a voice closer to the original song.

How can this be done?:

  1. We could first send the original lyrics of the song in question to SUNO's AI.

  2. Then we could send the .mp3 file with the song and vocals.

  3. And finally, send the .mp3 file with just the instrumental.

With these 3 pieces of information, SUNO's AI would analyze and create a song with a structure as close as possible to the original based on the new lyrics that will be sent, which will have to follow the same structure as the original. However, the important point of this is that it would have to be the entire .mp3 files, and not just 60 seconds at most.

nova flicker
jolly wharf
#

Implement a STICH or MERGE function, whereby we select two songs from our song list and combine them.

Also the ability to add to the intro of a song would be great!

Up vote please.

nova flicker
clever briar
#

Make the Covers feature free to all users

sweet flicker
#

Suno showcase suggestion Horizontal Discord links players and restriction to 1 link per person. (or 1 embedding per person) Some people will link 4 links right after your post and your post gets pushed away. It also makes me not even want to listen to the 4 links poster. And that's a shame cause I'm sure there will be some good songs in there.

magic granite
#

Cover needs to major work in detecting and implementing covers that blend or converts genre. I am having a MAJOR issue with turning the main flow of a violin sample into something plucking based (guitar, lute, etc). It's not even attempting to pick up the rhythm and tosses the entire Cover sample out the window. Zero recognition of the source.

Edit: This is after the result of hundreds of Cover attempts. 90% of all Covers have come out completely in ignorance of the Cover sample provided.

prime rock
rustic radish
rustic radish
sage mirage
#

It fundamentally is how Suno works is all, Master only. Trying to train on stems alone might be a practical impossibility in that there is a lot more full music to train on rather than stems

#

There might not be enough stems to train, moreover to keep it coherent between different stems

rustic radish
#

Excuses don't make my workflow any better. It's possible. Fact

tepid trench
#

It currently is not possible.

twin grove
#

so, how is it deciding what melodies to play? It's generating notation somewhere along the line

#

anyway... my suggestion: Filter by Disliked in Library, for ease in deletion

tepid folio
#

I think a new feature that would be fairly easy to implement is the ability to save prompt presets

Would love the ability to choose from a list of presets we save instead of remembering sets or pasting from a list manually, especially with the new negative prompts

sage mirage
#

If it were easy, we'd have it.

royal venture
# rustic radish Right, but this could be implemented. Don't tell me it's not possible or not ver...

Im sorry but you are simply wrong. Suno doesn't use stems, it doesn't make stems. Proper stems requires that every instrument and each individual vocal are generated separately, and all individual generations are synced and know what is happening in the other stems, in order for them to line up properly, make sense, and sound good in the full context of the song. Is it possible, maybe. But the cost in cpu power would grow exponentially for a single user, let alone a million users who can use up to 10k credits throughout a month, not counting paid alternate accounts, or credit top ups. The increase in server capacity and compute alone would sky rocket, would require an equal increase in cost for generations, and that completely excludes the complex task of somehow trying to get maybe up to 10 separate stems to generate on the fly in unison with other stems to make a coherent song.

Suno does not use samples in your generations. Maybe they reference samples, because they had to train AI somehow, but nothing in what you hear has samples in it. If it does the song is probably automatically getting deleted, and playback will stop even if you are still listening to it when it happens.

robust jewel
#

My understanding of how LLMs work (and someone correct me if this is wrong) is that they take the input and process is, changing values internally to determine what the most reasonable next single step might be, then they pick one of the highest rated possibilities. With that single step decided, they append that to the input and run things again, building up the output bit-by-bit and running it again as further input. If that is how they work, then, no, there is at no point any notation, even internally. It is only ever "what has been done", which is the thing it outputs to the user.

frozen nacelle
#

Dear developers. Please make it possible to choose your voice

iron parcel
#

Hello, I create a song, it always show error: "An error occurred." Why? Where do IT of Suno AI, please fix or everyone is layoff, so no one know fix this error :))))))))))

image.png
robust jewel
#

It's generating things just fine right now.

#

It's possible there is something getting flagged with your prompt, though.

fallen cloud
#

When will the streaming platform be launched? Put it on Google Store or Apple Store like spotify.

edgy wren
#

GM - does anyone have experience manually taking the songs and making stems in a DAW?

past plover
#

HELLO Suno!

I love it. Best app ever. And pricing is great. THANK YOU 💕

I would love to give from 1 to 5 stars to my songs. Only like or dislike or nothing is not enough.

Stars or points would help me later to check the song.

And I would love to give personal info or comment to my songs. Just small box to make few notes about the song.

Are these ideas something others would like to have..

modest frigate
#

most of these suggestions and requests you give here recently for the user interaface and song management have been made already 6 months ago (just look up in the thread), but nothing happens... And most will never be implemented....or maybe only in 2030 if they do in this rhythm... like a minor change in 3-4 weeks

delicate cedar
modest frigate
#

ALL these suggestions regarding UI and song management have been made in the last 6 months can be implemented in a single month by a good UI programmer or by 2-3 randomly chosen programmers from Fivver, given a right project manager and a right software spec. If Suno cannot do with a lot of money and in 6 months, what can be done by anybody else in few days or weeks, then Suno has a big problem. Some of us even offered to help for free, but no response from Suno whatsoever. As I said before: we are talking to black hole that pretend to be alive but it's not 🙂

delicate cedar
#

🥱

#

Then apply.

#

lol

#

Be the meme.

modest frigate
#

every time I express some critics you offer me a job 🙂 But I dont need a job. And even if I would apply as a UI project manager, who do you think will give me this job from Suno where they don't give even an answer here or via email? Nobody. Cause anybody there is probably overwhelmed with something.

delicate cedar
#

It's easy to criticize. A bit harder to actually fix things. So I mean...if we've all been suggesting you apply maybe you should. Like I said...be the meme.

modest frigate
#

I applied already 5 months ago !

#

cause I realized that maybe Suno needs help and manpower...

#

but not because I needed a job, but cause I wanted to have these issues fixed

ebon kindle
#

I’ve used Suno for a few weeks and I love it. With its Covers beta feature, I believe it has the potential to completely redefine music production as we know it, as long as it can keep innovating and adapting to the tools music producers currently use.

As far as I am concerned, here are the feature requests I would love to pay for (in a special Music Pro subscription package, for instance):

  • ability to set a track duration (particularly useful for video soundtracks)
  • ability to export the produced song as MIDI file (I may be dreaming here, but that would be an awesome one)
  • ability to separate the instrumental and the vocal tracks from the produced song and fine tune the lyrics to regenerate the vocal track (without regenerating the instrumental track)
  • ability to specify the song structure (intro, number of verses, chorus, bridge, outro)
  • ability to assign parts of the Cover input track to the verse, the pre-chorus, the chorus, the bridge, etc..

In essence, the idea is to provide a bit more control over how the Suno AI generates songs. Covers already gives control over the song melody, and that’s a great first step. But if Suno wants to be a breakthrough in the music industry, there a few more features it must implement.

I hope there’s a PM tracking the user feedback here 😉

lavish sleet
#

A few straightforward features: the ability to organize your playlists in a custom order and the option to sort songs and playlists by view count.

wanton rover
#

As a paid Suno user, I’d like to suggest a feature that could enhance the platform for those of us focused on mixing and mastering after making a song or album Specifically so that it can be submitted to distributions like CD baby, Tunecore and Distrokid, it would be great to have:

A User Friendly Interface and Custom Presets for Different Genres Of Music Mixing and Mastering: Including detailed EQ, compression, and reverb settings tailored to common vocal and instrumental styles.

Dynamic Vocal Processing Tools: Like multi-band compression, advanced saturation, and harmonic exciters specifically designed to add character to vocals.

AI-Assisted Mastering: An AI tool that could analyze my track and suggest mastering levels based on genre-specific norms, allowing for adjustments but starting with a solid baseline.

I believe these features would add immense value for users working in genres like Hip-Hop, Rock, R&b and Country, who need precise control over their mix and master.

It would be nice to implement something like this for paid users without having to leave the site maybe charge a Suno as you go plan lol that's affordable for users who just want to mix a few songs or add a $20/mo mixing and mastering subscription plan to pro and premium users?

Thank you for considering this!

wanton rover
brazen ether
#

You know what Suno should be adding? Additional adherence to chords. Like user gets to input Chord progression, and song generated will follow it. In that way, more originality or more input from the user can be done. And also adherence to BPM. so users could also set the bpm that they want for their song.

hearty ivy
#

Hello, sorry if I’m posting this in the wrong place but how and where do I suggest to add more genre. I’m Haitian and we do have a lot of genre in our culture. None are present and coming up with the right rhythm. How can this be fixed? How can I help? For example, we have some genres that are uptempo, there are no such options. Thanks!

tidal panther
#

Can we download playlists please?

marsh lichen
pastel finch
#

PLEASE add an option for a midi export for paid subscribers. This would make it so much more convenient

pulsar bone
pulsar bone
long scroll
#

Tell the user during flagged lyrics what is causing the song to be flagged and not produced.

#

It seemingly doesn't make sense how it's setup, I'm just guessing at random words hoping it's one of the 300

#

The word sacrifice isn't allowed?

#

That's literal NPC logic

wanton rover
turbid prairie
#

we need covers in free accounts tbh

#

just sayin

chilly cradle
#

yo maybe not have so much weird moderation when it comes to prompts. stop censoring what my lyrics are. im trying to make a song with fl studio and i dont have anyone to rap what i have for me, or have a good place to rap my lyrics because they are explicit. they are loud. thats why i need the robot to sing for me for my song.

#

its not even politicalor anything, its just dark lyrics. why must this be censored?

#

it just sort of sucks:/

#

it takes away from creativity from people like me who use ai ussually only for singing/rapping cause not everyone is rich to pay someone else to rap it for them,and suno can make basically anything sound good so like

#

come on

tiny carbon
turbid prairie
tiny carbon
#

Yeah same

void spire
#

Would be awesome if you develop a „change“ feature, where you could change out single words or parts of a song or expand a song in the middle (like add a guitar solo or record scratches with prompts). I mostly make German songs with Suno and often some German words are pronounced English and covering and changing the text mostly completely switches styles or doesn’t do anything. Would be awesome! I highly appreciate your work, keep up being awesome 🙂

modest frigate
#

they will, I guess "inpaint" (Reparatur) is the next audio feature to be implemented

valid yarrow
#

They should add a lock you can turn on, so you cant delete a song by mistake.

marsh lichen
clever briar
twin grove
long scroll
#

Bro I removed a space and it went back to flagging other random shit

#

💀

clever harness
#

Might be nice to alter color of genres when limit reached

It might be nice to offer visual feedback to show I can no longer add a genre due to hitting the max limit. Bonus points for whittling down the list to only genres that fit the remaining characters when possible. 😄

Screenshot_2024-10-08_at_6.44.41_PM.png
zealous musk
long scroll
#

XD

ebon kindle
delicate cedar
jagged fable
#

There needs to be more user control for the cover feature, as of now it is beyond unreliable. One of four things happens and there is not much of a pattern I can pick up on:

knightno It may slightly change the song's key (usually up or down a step or 2) and does nothing else at all. This is pretty common and a waste of credits.

knightno It may produce a result that is completely unrecognizable from the audio it is covering, as if it didn't even reference it.

knightno It may do nothing at all and add a few artifacts to the original audio

knightyes It may work correctly and listen to the prompt and input audio, creatively modifying it and bending the genre without sounding too different. I feel this is rarer than it should be.

I don't want to have to use your competitor as an example, but Udio has a sliding scale where you can adjust the "similarity" of the cover (which they call remix). You have complete control over how similar the output is and can hone in on a sweet spot based off your preference. Please consider adding something like this.

tawny wasp
#

I have a MAJOR complaint. Credits that I had paid for on a prior billing plan have dissapeared. nearly 2000 credits gone. I had previously unsubscribed from the plan, but as I have PAID for those credits I should get to keep them.

delicate cedar
tawny wasp
#

I am arealy aware

delicate cedar
#

Read the print on the bottom of the page.

tawny wasp
#

I am complaining specifically about that lol

delicate cedar
#

Oh. If you knew already then what is the issue?

tawny wasp
#

that is the issue

delicate cedar
#

🤦‍♂️

tawny wasp
#

(specifically my reccomendation here is that credits that have been paid for should be kept period.)

delicate cedar
#

I should get a hold of Netflix and demand that I get a free month since I didn't watch it when I had it.

tawny wasp
#

lol

#

vastly different thing here but please go on

#

im not asking for more points, im asking to keep the points that I paid for

delicate cedar
#

Oh and the hamburger joint down the street owes me money. I ate the burger. It was good but not as good as I think it should have been so I want my money back.

tawny wasp
#

classic discord moderator moment

#

do you just make it a hobby to go "um akshually" in this channel?

delicate cedar
#

You agreed to the terms and used the service. Now you want to post facto alter the terms which provided you the service.

tawny wasp
#

"how dare you reccomend a feature in a channel dedicated to reccomending features"

delicate cedar
#

I'm not sure how this level of miscommunication exists to be honest.

tawny wasp
#

Its not miscommunication, youre just being a redditor

#

you see someone go "hey I would like to keep the thing I already paid for, im not asking for more, im asking to keep what I already have" and proceeding to exaggerate it as if I am asking for endless points

#

ive seen you act like this a lot, and its honestly suprising youre still a mod with the disrespect you show to the members here

delicate cedar
#

You paid for a subscription set to a term = access. That access was metered = credits.

If you didn't use the credits for that term that's on you. Not the service.

tawny wasp
#

what part of suggestion are you struggling to understand

jagged fable
tawny wasp
#

Its like if you bought a gallon of milk and you didnt finish it within a week so I came into your home and took the rest of it

delicate cedar
#

The part where the suggestion is basically 'give me unlimited access after my subscription ended' is where I get confused.

tawny wasp
#

called it

#

maybe gain a little bit of reading comprehension

#

I am not asking for a top up back to the 2500, I am asking to keep the points I havent yet used

#

which I was pretty clear about

#

but acknowledging that would ruin any point you have, and you wouldnt have an opportunity to feel powerful

delicate cedar
# tawny wasp Its like if you bought a gallon of milk and you didnt finish it within a week so...

I actually have a gallon of milk that "expired" September 22nd. It's still been good. so I continue to check it and use it. If it goes bad tomorrow the fact that there is still more milk in the gallon doesn't entitle me to go back to the store and demand a replacement. I knew when I got the milk that there was a limited time for me to consume the milk. It's on me if I don't use it before it goes bad.

tawny wasp
#

except in this case I would take the gallon on that date, milk good or not

#

thank you for understanding my point

#

please continue to pretend the situation is different

#

"Sorry, I know the milk is still good, but the expiry date (tos) is clear, you no longer have access to this milk"

#

Ill say it once more, i am NOT asking for additional credits, I am asking to keep the credits I already have

delicate cedar
#

The situation is different. The "milk" (the credits) have expired. I don't know how to make it any clearer.

#

Have a pleasant evening.

tawny wasp
#

So then go dump out that gallon of milk since it expired lol

#

Also I love how when your analogy fell apart you changed to "the situation is different"

#

and also continue to pretend im demanding more credits, instead of oh idk, keep the credits I already had.

half escarp
#

This is idk here i would like it so i can not only crop my songs to sections but conentate them into a new sample without having to reimport or use 3 rd party tools

#

so think merge 10 second samples etc

#

like 2 or more into new sample system

#

i already do it but it would be nice not to reimport since the system identify its own vocals as voices

#

so then i cant release on suno cause its restricted

#

ughhh liimitation

#

s

topaz eagle
#

Damn I was thinking... A great function would be having an addon to upload a midi for the vocal line to sing.

As an example... We making a leadmelodie but mostly the ai uses its own melody. If we could make the ai-vocal to special midisnips as example for the prompts:
Verse,
Chorus,
Bridge,
And else... It would be very nice

turbid wraith
#

R & B would be nice

versed inlet
#

Export MIDI for paid subs needs to happen ASAP

Stems just aren't what they're advertised to be.
Vocals aren't isolated well enough.

hard tinsel
#

We need that and the team should focus only on this from now on and also to maximize the output quality which is poor..

clever briar
next mortar
#

I'm sure I'm not the only one to suggest this... But a way to edit a song lyrics or prompt without disrupting the song itself. Sometimes you have something extraordinary and you have 1 or 2 words wrong or maybe Suno sings a prompt. It would be nice to be able to just change that up.

modest frigate
#

if Suno still wants to stay "parameter free" and "baby-easy to use", so only a button press would start an action, it should split the COVER function
into 2 different commands: "Cover" and "Remake".

  • Cover should try to do a different version of the song.
  • Remake should try to do an identical remake of the song (in order to fix bad audio quality, glitches, bad pronunciation etc.)

The other option is to place a slider/parameter to let us choose what we want.
A slider on the waveform to decide what you don't want to cover (like the intro for example) it would be also great.

Ah, is there any forecast for the Cover function getting out of the "beta" status?
Or what has been improved already in the 15 days passed so far with this function?

lunar topaz
#

hello, I thought I posted 2-3 point suggestion in here, but doesn't seem to have appeared

#

anyway: suggestion : syntax highlighting (API style) primarily parenthesis [ and ( . ie you hover on one and suno auto highlights it's matching pair (eg bright purple) (or mismatched pair) (eg bright yellow) or at very least a quick scan of the input and a flag notifying you of missing parenthesis. that would be great, ill post the rest another time

wind crest
#

thrilled to see you're adding the ability to replace sections (even if just for pro) but i see that it's for sections 10-30 seconds long. in future i hope you will let us just hilight any portion of the song even if it's 1 second or less, because that's all that's needed to replace a single garbled or mispronounced word. generally i don;t want to replace whole large sections, only the one moment that ruins an otherwise perfect section

jolly wharf
#

Ability to MERGE songs.

Please up vote.

marsh lichen
wind crest
#

how does it currently work, with timestamps?

#

to be fully useful it'll need to work with either extremely granular timestamps that go into miliseconds, or with hilighting a section of a visualised song, since a naff section never conveniently starts and ends on a second

#

i can't wait to go back and correct some bad words in songs i use for language practice/vocab learning that are otherwise perfect

#

i would pay for it if the tool is adjusted to be be able to edit very small and very specific sections

stray urchin
#

The new feature should allow you to put any unwanted clips in the trash from the editor window.

slow wave
#

I believe it would be better to generate the stems in isolation and keep them seperated, rather than just using AI to seperate the stems. There is lot of unwanted artifacts in the sound.

#

One feature I would really love to see is the ability to generate vocals without instruments. When I seen the stems feature I was ecstatic, but then after trying it out, I was disappointed as I realized it was just the same mediocre stems seperation feature that many other services already do.

tawny wasp
#

Is there a way to see who is following us? Would be nice to see to make it easier to follow them back.

still parcel
tawny wasp
#

oh perfect thank you!

void spire
#

A reuse voice or reuse style feature would be awesome, another great think would be the option to use your own instrumental

twin grove
delicate cedar
twin grove
#

yes, immediately

delicate cedar
#

That sounds like a bug. That's different.

#

Let me kick this up. Thanks for clarifiying!

twin grove
#

so, with the gallon of milk analogy, you decide you don't want to buy any more of it, but then the grocer comes and dumps out what you haven't used

#

or, more technically, it's like the 'paid sub' flag is unset immediately, rather than on the end date

grave salmon
#

DM if you have questions or need help!

scarlet shell
#

you know how when you click get full song it takes all the clips and puts them together it would be cool if you could upload your own 2 audio files and it puts them together

twin grove
#

wasn't me doing the unsubbing, but I've seen many people that have canceled and had their sub status and credits disappear instantly, rather than the end of their month

tawny wasp
tawny wasp
#

also I think its still a pro/premiere only feature

#

(also it may be more for the melody rather than the voice, of which I would love to have a feature for that)

mortal fern
#

Followers notify is awful. I don’t want to know the last one+25 more… i want to know who are the other 25!

tawny wasp
#

it would however be nice to see if there is an "expand message" feature to see who all subscribed with that notificiation

#

(unless there already is lol, i dont regularly get subscribers, just responding to a cool idea that was proposed)

clear copper
#

Suggestion: Option for "compliment" song in the same area where you can report bugs/innapropriate. It could work like this: you select it, you press check the boxes with the things you like about the generated song and/or you write what you liked about the song, then the AI will try to focus more on the details you liked about the songs.

potent rain
#

Why is the replace section selection limited to 30 seconds? I want to bridge two distant parts of a song but I don't need to replace the entire length, the new section can be 30 seconds long

modest frigate
#

thanks for the REPLACE function Suno! I have not used it yet, but it is done smartly (much better than Udio!).
I've just watched the tutorial video from Franklin and observed his credits 🙂 He does not seem to have much time to spend those 🙂
I hope the dev who implemented this feature will have some time soon to add also fading out to the crop function.

image.png
glass barn
#

Let putting Prompts into an extended song actually do something

The song could be super super happy and then I give it all these different dark Prompts for the extended version and it’s still super super happy with nothing I asked for

north jasper
#

I haven't tried the new Replace section at lenght yet, but 10 seconds gives enough room for the engine to change the tone of the section it seems. Is there a way it could match more what the original tone was... Many songs i have that i really enjoy just have a few words that need fixing. In either case, my mouth dropped when the announcement came for the new feauture. Thank you SUNO!

rancid saddle
#

i think we need to do voice cloning for the singing, so we can brand songs to our voice

formal island
#

The lyrics sometimes start very abruptly in the beginning, leaving the song feel rough/introless, and it would be nice if we could slide over the current song over to insert an intro before lyrics start, or to add some space at the end for a dedicated musical outro. It ignores meta tags and instructions frequently, and I have some great songs that really just need a measure or two to spin up before vocals. 🙂

formal island
desert nexus
#

no matter how hard I try to replace lyrics in a song with just instrumental or just straight up remove the original lyric it will not work. Can we add an instrumental toggle for replacing to force it to not generate any singing.

Also any chance to get slightly shorter replace options? Sometimes needing to replace 10 secs can cut into sections that you don't want to change a bit too much. That or we need a way to specify more about what we want to change and what we want to maintain from the section itself. I know it recreates everything each time in that section but if it could try to recreate vocals while changing music or change vocals while keeping music intact etc Generally speaking I think most people will want more control over aiming the model at the instruments or vocals but normally not both at the same time.

Also limits are very frustrating to work around. For uploads (2 mins - I know its more than it used to be) and replace sections can only be done on songs under 4 mins.

A lot of the time I take stuff out of suno and bring it into my daw and currently there is no way to reupload it back into suno to continue working on it if its over 2 minutes. Also some songs that are like 4:20 seconds in my library i have to take out cut up into 2 min sections reupload to replace parts and then stitch it all back together manually and all without an option to then reupload it back into suno since im stuck with a 2 min upload... i think you see the issues.

stone steppe
#

The ability to create sound effects, directly.

  • a separate create page for SFX, similar to elevenlabs, where the user can make specific sound effects for their music compositions. It would be very cool if we could adjust the length of how long the clip should be. To avoid it being confused as a music generator, it should have a max of 50 seconds.
#

I love the idea of uploading a kick sample to suno and prompting it to make 10 variations based on that sample.

#

Stable Audio is doing this right now, so it is possible.

sonic cypress
#

Replace section available for songs over 4 minutes would be great.

unborn token
#

Just came to say the exact same thing. got older songs I'd love to change one word in, but most over 4 minutes

topaz eagle
#

Hey another idea from hell:
How about to create own Vocals with the very own Voice.
I mean i like the voices from SunoAI but i would like to hear myself. Just to prevent any singer to say: "Hey that's my voice!"

That would be a gigantic feature!

atomic steeple
#

When searching for songs in your library there should be a button to filter for the "Full song" tag.

weary spindle
#

with replace section being available, the only feature missing now is folders to organize songs... 🙂

sour geyser
#

Suggestion: pay for individual tracks (made in the free plan) to unlock usage rights.
I’ve got a few tracks I made in the FREE mode that I’d like to use in a monetized video. I know the rules don’t allow that, but I’m willing to pay to get the rights for the music. Wouldn’t Suno only benefit from allowing payment for tracks we made earlier (in the free plan) so we can use some of those that turned out really well? Or does that option already exist and I’m just missing it?

#

Or is there a way to remake the exact same track (without changing a thing, exactly the same) while I’m on the paid plan?

tepid trench
marsh lichen
stone steppe
marsh lichen
umbral grove
#

comment section would be nice where people can comment and let people know what they think (be for all users though)

modest frigate
#

I used once "reverb" for a vocal and it worked right away... I guess the model knows most of the sound effects like echo, hall, reverb, vocoder, flanger etc

royal hull
#

idk Suno guys, maybe focus your efforts on one feature at a time until it's at least stable enough to use? covers still work like garbage, and now we have more barely working stuff. making hundreds of gens is starting to get me frustrated.

rain barn
#

Hello there! I'd like to leave three suggestions:

The first one is to make the song list in the "create" section longer, so I don't have to go to the specific tab to see everything. Sometimes I want to check something from earlier, but I can't because it only shows a few songs in the created songs list.

The second is: I'm a singer, and I'm excited to use Suno to arrange my songs. I noticed that when I sing a cappella, it doesn't always get the chords I envisioned for that part. Maybe in the "replace" section, there could be an option to change specific chords. Of course, it would have to show in real-time what's being played where, so I could choose which chord to play. It could give suggestions, when clicking on the chord, of the "possible" chords that logically fit that part. Or, it could leave it open for you to choose any other chord. When clicking on the chord, it could be possible to hear it, so you could be sure. And being able to hear the new chord with the sung vocals at the same time would be amazing.

The third suggestion is in the "replace" section, where we could choose whether we want to change the lyrics or the melody. I tested the function to replace lyrics, but it's not 100% there yet. (I even recorded a video for my channel testing this function, and it's still a bit frustrating… but I know it's very close, and I'm really excited for when it gets really good).

Other than that, thank you so much for all the improvements. Suno is getting more amazing and remains my top 1 most used AI for music.

Thanks!

modest frigate
#

LYRICS WINDOW: make, if possible, the lyrics window a bit longer, vertically and also add a "hovering" window copy of it that you can open (and close) to see the whole lyrics if you want to (so we dont need extra text editors)
CREATE WINDOW LIST: make either the list longer or best: make 2-3 pages of this song list with tabs for each of them, so you can go deeper in the recent songs.
CREATE WINDOW FILTER: add a filter to select what you show in the create song list. Place this filter at the bottom, there is the best spot. So even if the song you want to edit is not there, you can filter it from the main folder/list (like Udio has).

elder talon
#

show what is the bad word in lyrics so i can change 1 letter from it

open pine
frosty breach
#

I would truly appreciate it if it were possible to for instance, have a preview of a part. Where you can listen to your latest extension with 5 or so seconds from the part prior, that aligns with what you just extended. To cut down on generating the entire 'whole' over and over again, just to hear if it aligns properly. To then continue on without even utilising the 'whole' I just created. It feels like both a waste on Suno's end, and time waster on the user's end.

signal shell
#

Hello! I have been having a great time with Suno AI since becoming a premium subscriber. The vocal emulation has been quite fantastic and frighteningly accurate! As a trained vocalist and professional musician, I am quite impressed. That all being said, I was hoping I could make a few suggestions - I am also requesting upvotes and reposts for this for anyone who would like to see these suggestions come to life, as my lone two-cents amount to very little without a consensus.

Some things that I had thought of:

-A select multiple option to delete many tracks at once, add to playlists, perhaps download multiple files or a zip
-Chord Progression and Scale Control via drop down or prompt - Could be based around Nashville Notation, Mood, Chords, or the option for all, but this level of control would be insane
-Optionality for vocal range and placement - crisp, airy, screaming, etc.

Now, I know that prompting can work for these, but I could see these types of options giving more consistency to the process. I think the application is amazing, but man I would really love to better integrate my knowledge of theory.

north jasper
#

Probably already mentioned it, but could we get Smart Playlists. This is a playlist that you could tie to search rules and populate with everything you generated for a given song based on song title, or perhaps combination of styles, or on a given time period. One of the challenges for premiere members is organization with so many generations.

signal shell
modest frigate
#

forget playlists, we need folders...

signal shell
north jasper
#

Folders would work too, also the ability to pick multiple songs at once and move them around.

signal shell
modest frigate
signal shell
north jasper
#

I know some things have come later, but i am actually impressed at the speeds things have been released, and pleased by the features they have released. The replace one is a game changer, it has worked for some of my songs but not others... The smart playlist idea is so that you don't have to sift through, and it can automatically organize things for you just like you would do on an email application when organizing emails.

unkempt berry
#

Permanently delete at once, like on the phone.

tulip fox
#

The option to remove the link to the original song from the cover song IF you are the creator of the original song.

I'll often regenerate a song as a cover to try and achieve better flow, inflections, and sound quality with the aim to replace the original.

tulip fox
wheat marsh
#

I think some of the suggestions I would give is a checkbox to "keep album art" of all extensions for part 2 and part 3 and so on and full song as it seems to get lost and if it was created on site I have to open up the image and resave it and upload it to the song which can be a hassel, also if your using an intro to extend from it always names the full song based on the intro thing so you have to rename it each time

wheat marsh
#

I would also suggest having a feature letting you know a songs already in a playlist as you can add the same one many times, that and adding multiple songs to a playlist at once using checkbox select or select all on page as playlist management is wonky.

#

oh and being able to see what playlists a song is already in so you dont have to guess(like selecting a song and it being like "in this and this" playlist)

desert swallow
#

I would like to suggest again that it would be nice to have a "stitch" function, where we can stitch two songs together in order to create a song medley, and have the AI intelligently segway between the two songs over an amount of time determined by the user. Thanks!

proud ingot
#

french suno

#

créer chanson arabe

near fern
#

this might change in future maybe, but so far AI generated content is not copyrighted content. Everbody can use it

#

that's said, I think it's good manner to respect such rules even if they have no legal purpose

tepid trench
near fern
#

true.
But I don't think they would ban a paying customer, just because the song he monetizes was generated BEFORE he start paying for the service

indigo ginkgo
#

Now that we have inpainting and cropping, the final thing we would need to be super-powerful in what we can do in music would be CONNECT. Where we can take two sections (possibly cropped) and connect them together. generate the in-between section.

This enables us to use the start of one gen and a end/middle section of another gen we love. How many times do you have a section you really love, but it's not in the one with the best vocals, or a solo you would love to have in that same song.

It would open up a lot of things creatively as well. I even think it would allow us to build loops, since you could do a CONNECT from something to that same something. Making it a loop-able piece... In my opinion, with crop, inpaint AND connect. You could build a career making AI music, because you would be able to do anything a client might want from you.

"Can you adjust this section?"
"Can we keep this solo, but only use the start of that?"
"Can you make this loop-able?"

So if I would have to vote on what feature I'd love next, it would be CONNECT, which connects one section to another.

modest frigate
#

We dont have yet FADING OUT, which is the most essential tool to end song.

#

MERGING or GETTING LOOPS FROM OTHER VERSIONS: what is missing and very useful is to be able to get a loops from other generation of the same song to place in the one you work at. Cause there are sometimes better choruses or intros or outros in other generations that won't repeat anymore and you will never get otherwise.
So lets name this function MERGE. You have the waveform of the destination song and you chose the insert point, you open another song (the source) with the waveform and mark the loop that you want to copy it /merge it in the destination.

crystal quartz
#

please fix the weird unusable vocals when singing over quiet sections. I assume it's effects that blend when over music but over silent parts it just makes it sound like AI and it's just not usable. I seriously have to throw away like 60% of songs because of this. How is everyone else dealing with it? just remaking? It's become a huge time sink

marsh lichen
arctic cave
#

Heyo, as a (very slowly) growing creator, I'd love to see a way to better protect my work. I've noticed that there is an option to "Reuse Prompt" and "Extend" on songs that are not your own, and I'd greatly appreciate it if there was a toggle to turn that off, making your content fully yours. It gives me anxiety thinking about others taking my lyrics and running with them.

Maybe some time in the future, I'd like to see what people do with my older songs once I'm bigger, but for now I hate the thought of people using the "Reuse Prompt" on my songs, especially because I write all of my own lyrics in every song that I write on Suno, so having my work protected would be amazing.

Perhaps you could do an option in the "Settings" for a "Disable By Default" checkbox for allowing others to cover/reuse promp/extend your music, and allow a setting to be turned on manually on a song-by-song basis if you want to allow that kind of remixing to specific songs at a later time.

modest wagon
#

Not sure if anyone has suggested this but what if each song had its own description box, that way if we need to, we could explain the meaning behind the song.

modest wagon
modest frigate
# marsh lichen Yes, vocals are AI-sounding very often - not only in quiet sections. That's why ...

Probably Suno will manage some day also to make the voice not sounding like AI anymore. Udio can do this already. But somehow I like much better the voices from Suno and the singing from Suno. Even if Udio can generate perfect natural singing many times, in 50-60% of the generations vocals sound like drunk vocals or like beginners would sing. I don't know how Suno managed to sing so well like in 99,99% of the cases and still sound a bit like AI (mostly only perceptible for profis though)
and Udio, despite such natural and perfect singing can produce in the same time so many unusable vocals. So I am waiting for the day when Suno will improve also the audio quality of the vocals. Cause the singing and the voices are already amazing!

buoyant smelt
#

make the cover feature free, I NEED IT BADLY!!!

clever briar
icy vortex
#

I suggest supporting all Unicode characters including Arabic ones especially in the case of bilingual lyrics:
#📚┃suno-school message

frail flicker
#

Please SUNO make chahging the album cover's for free user's to, it's kinda dumb to be honest that only priemium users can do it.

robust jewel
#

At the very least, it would be nice if it went back to how it use to be and we could pick any of the images generated in the chain of extensions. Use to be able to pick the image for part one to be the image for the whole thing, which was often more accurate to the idea of the song. Part one probably has 90+% of your lyrics in it and part two has the outro, often as not, but you get stuck with whatever was generated for the outro. Not sure why that changed a little while back, but it was a negative one.

arctic cave
# robust jewel At the *very* least, it would be nice if it went back to how it use to be and we...

I know, I really miss that!

I remember being able to just choose rq from the available covers, but now you either have to go back and save the original cover off of the original generation that you chose to then apply it to your finished product which can be a hassle, or you can do another AI image prompt (which usually isn't top-tier on Suno and best done on 3rd-party websites) which is also a hassle to do.

#

I just use Perchance for my covers now, but it at least gets the job done.

robust jewel
#

I'd like to have a prompt box for cover image, really, but at least being able to pick from other images in the same chain was better than getting a weird bit of line art because the ending music was cut off and had to be regenerated with no lyrics.

1.jpeg
arctic cave
zenith rivet
rocky wigeon
#

Simple suggestion for those of us who may be memory conscious of servers that we don't maintain - a function to COMPLETELY purge our trash section without having to go through individually perma-deleting each generation. For me personally, if it's going in trash, that is exactly what it is. I'm not gonna dumpster dive for an old generation that I didn't like to begin with.

spiral vessel
#

I noticed Suno doesn't seem to be trained on certain instruments: cowbell, theremin, triangle. It's been hard to get it to do harmonica. So my suggestion is, train on more instruments 🙂

keen burrow
#

1️⃣ It'd be great to be able to just delete generations directly in the Replace Section screen on the web (from the three dot menu)
2️⃣ It would be good if the AI actually followed the lyric changes in the Replace Section (thankfully, they're free, but gees, what a time waster)
3️⃣ An option to keep the vocal pattern the same in the Replace Section would be helpful (sometimes they have slight errors and need to be redone)
4️⃣ It would be good to have a way of deleting multiple files at once from the Library pages
5️⃣ A way to group generations within the Library would clean it up sooo much (folders or something) and save loads of time
6️⃣ The ability to clone an uploaded singing voice would be awesome

🩵

strong horizon
#

Hello 🙂 do you think that it will soon be possible to separate the different tracks of the songs (Guitare/drum/keyboard...) ?

lavish edge
#

Can I delete all My thrash can content definitively or I really have to delete The content of thrash one by one?

ripe tusk
#

Can you please make a way to keep the same singer/instruments and arrangement but tweak the lyrics and regenerate it. I use Suno as a demo maker, feed it my own music and lyrics and to get the song where I hear it should be. I get an almost perfect version but want to change a word or two. If I make the tweaks and regenerate it takes me away on a tangent and waste a lot of credits in the pursuit of getting it right. Please, please give us an edit facility.

hollow locust
#

Please fix it constantly elongating tthheeeeee wrong word.
Which honestly sounds better:
To theeeeessse distant peaceful seas,
To these distant peaceful seas.
?

The key words are 'distant' and 'peaceful' not, these. Why would it think "these" is an important word?

kindred cypress
#

#💡┃suggestions-web Hi, guys please change "replace section" limits... for now it is available only for songs less then 4 minutes. i was making song from different parts, and now when it is ready, but i need to six some lyrics, i can't coz of that limit.

#

p.s. song cover doesn't make covers over 3min and 17 seconds

#

can it be fixed

obsidian wolf
#

It would be neat if we could crop the Replace Selection audio clips

jagged chasm
#

The menu that you open when click on the three dots on the right of each song doesn't have a "Song details" button.

The missing button is present in the menu that is on the right.

image.png image.png
stone steppe
#

Add a direct "outro" option to force the song to end with v3.5

Thank you

twin grove
#

We need a better way to ensure that Replace Section honors the original melody. Beyond being flat-out broken, and 100% useless for fixing anything beyond 2 minutes in, if you try to make any real changes, it's like an Extension, and that's not what were wanting, especially when it turns a pre-chorus with a nice 'wave-like' melodic structure into flat, boring mess

tidal zinc
#

Is it so hard to teach AI to understand what sibilances are? That's a pain, no miracle command makes it understand...

#

The replacement section is not 100%, I confirm and the system keeps generating the full version forever, so I can no longer extend it until it loads

ionic axle
#

Can we get a feature to ADD lyrics to an instrumental track?

#

The Stem feature removing seperating lyrics from instrumental isn't very good, so I'd rather take the path of creating a good instrumental beat and adding vocals to it

marsh lichen
ionic axle
marsh lichen
opal geode
#

I have a couple of suggestions for future SUNO releases

  1. Give Suno the ability to WRITE out the music and the different parts. That way, it can be reproduced by a band. Lets say, someone hears your music on Spotify. They contact you and say they want to license the music for their band. They will need the written music ( or have someone take the expensive and time-consuming route of transcribing the music from the audio - there are people who do that). If SUNO could do it, WHOA Nelly, it opens up endless possibilities for licensing the music besides just elevator music. Every songwriter (even ones who cant read music) eventually need to have their music written out. There is a whole industry of people who transcribe music into written form based on just guitar chords.
  2. Have a section (like the lyrics section, or style section) where you describe the structure of the piece (ie ABCBD, ABCD, etc). It can also be done with metatags as long as SUNO understands the tags and doesn't ignore them. This opens up many more combinations of songs. Right now, the default structure is ABAB Outro (maybe a bridge), which is standard pop and rock music structure.
blissful pendant
#

Can someone give me the size for the videos Thanks

#

Also how do I edit the lyrics and redo video after making a song

woven fractal
modest frigate
#

FAVORITE TAGS: Place at least a button near the style prompt window that you can press to paste in the window your favorite tags.
So we don't have to type them every time like in the MS DOS time. Best would be to have more buttons cause we make different music.
Like F1, F2, F3. And that tags should be saved in the preferences window (that you still need to implement in the future).
And by the way, the future is now, we live in 2024 and Musk even made the "terminator", so we
don't have to type everything, right? Even AI has been invented in the meantime! Geez!
Until you make the UI the way it should be and any other company does, we will probably all be killed by the
Musks Terminator. So please hurry up a bit, maybe we can tame the Terminators with some music 🙂

image.png
#

COVER is the biggest credits and time waster: since you did not manage to fix the cover or at least gives us a parameter to control it somehow, you should
lower the cost for the credits (5 instead of 10). This would be fair.
Cause it never does what you want! It makes only clones when you want smth else, and only covers when you want a remix.
As it would know that and would like to tease you 🙂

amber violet
#

PLEASE! Please! please... allow me to create songs (and all edits, etc) within a folder. I hate moving them all manually. The way it is, I have to almost work very linearly on my songs.

leaden cradle
# modest frigate COVER is the biggest credits and time waster: since you did not manage to fix th...

My biggest gripe with 'cover' is that it never goes the distance. The 'cover' is just cut off and never manages to finish the track. It also seem to have a majority of the covers set to 3:22 for some reason. Its also too costly for what it does and for all the issues associated with the feature.

For subscribers, there should be far more options to let the AI know what we want it to do and how to do it as far as covers go, especially since it cost a full 10 creds to use.

modest frigate
#

yes, I even forgot about the 3:24 limit. And also that cannot cover a whole song uploaded

signal shell
#

A good feature for those using their own voice would be an option to give Suno feedback. Is this closer to your voice? Yes or no. It gave me an immediately perfect generation with my newest recording from my AT2020 and then afterwards it started getting farther away from my true voice...not terribly far, but far enough for us to notice... Another feedback I would like to give it is "Is this how you would use your voice?" Feedback. I don't know why it makes me sing like a mumble rapper sometimes...or super breathy - no indicator of that was given

half escarp
#

Can we have a part of this that is made for daw users i kinda am bored with inferace controls would prefer a system where actual control come in

#

technically it would require us to contol learning level of this but idk i cant fooog with suno so

#

well i can but its old suno so

opal night
#

Binaural Beats 🙌

sterile stream
#

I love the new notifications!

I would love to be able to comment / review / rate the creations from other users, chat with other creators, publish posts or stories to call the attention of my followers to a new release, etc.

We do have the "trending" pages, but how about daily refreshed charts? Radios?

Although the focus is on creating music (and I understand it), I think that most of the users want their creations to be heard. We all have a masterpiece that we believe deserves to be known by a larger audience. 🙂

tough kite
#

new suggestion a way to make multiple songs in one go even maybe make an album from the said songs another suggestion under the create tab is to be able put things in folders for easy sorting

lapis osprey
#

Will you please add the ability to bulk download? It is extremely time consuming to download many songs one-by-one.

worn atlas
#

I am just testing the Replace feature and it is not helping me a lot. I got two new issues with the process.
My song was pretty good with a 4m16s length.
Only one line in my first Verse was to be modified..

  1. The length was an issue. Suno does not accept more than 4mn for the length. So the previous feature was necessary to crop the song to less than 4mn.
    2)Suno does not allow me to replace a single line. But if i add additionnal lines in the édition suno is singing these lines differently. So suno solves perfectly my problem but introduces a new one.
    In conclusion i got the original issue solved, but two new issues introduced.
stone solar
#

small suggestion: adding fade options with crop would be nice

gritty prism
#

...but I would (in addition to the above) like to see the overall server load, because the results when there are a lot of people hanging out are the most depressing... here you have noise, and confusion in words, and in the structure of songs. Although - why am I telling you - you've all been through this. And so - seeing that the server is overloaded - I'll calmly wait...

amber violet
#

For the Edit Replace feature, it would be great to have a sliders that focus on changing something or not and how much (like keep the melody, change instruments, keep the length, etc).

edgy seal
#

clicking link in showcase to like the song makes me have to sign in to a new suno not the one I have open already

worn atlas
unkempt shard
#

Hi.
Is it possible to make it so that Root Key and bpm are displayed next to the track description?

unkempt shard
#

P. S.
And this will be an excellent improvement if Suno includes bpm info in the final wav file. 🙂

odd reef
#

It would be grand if one could empty one's entire trash with one click.

subtle yacht
#

It would be great generate a short video clip for social max 1min

unkempt shard
spice fog
spark magnet
#

An actual radio network covering most genres

raw mist
#

As a paid user, looking forward to when Suno can get clearer vocals.

indigo ginkgo
#

Hey Suno,

Me again, is there any way you could up the trending slots from 35 to 60? It would allow so many more people to trend, more songs to be featured and be generally better for the community.

I truly cannot see a downside,

Thank you for coming to my ted-talk 😉 ,

ps. I'd also love the ability to gift someone some credits or 1-month suno membership

analog egret
indigo ginkgo
uncut otter
#

More sorting methods. Right now power users who generate a ton are negatively impacted, there are no smart ways to search for stuff. A couple suggestions;

  1. automatic genre grouping by tag
  2. only show disliked (for easy deletion)
  3. foldering of generations that are covers/continuations of an uploaded snippet
  4. a page system that is reflected in the browser URL
manic cypress
#

Cropping needs to zoom in so we have a better control of cropping

untold flume
#

can we get THX 7.1 dolby kinds of creations possible so i can make bees fly around my headset...

long bloom
#

once again (it's been about 6 mo) I'd like to suggest a version of suno that runs on one's own hardware. That could be running in webgpu in a web app for example, and it could be only available to pro users. I would love to put my gpu to work and not waste credits

#

It would reduce server load as well

clear mulch
#

I think the best great feature we all need is some "Virtual band/artist maker", where you choose a track and then can create entire new songs using the same vocalist(s), instruments, style of singing, guitar tunes or song elements.

tulip sedge
#

a trim feature would be nice. I often have dead space at the end of songs.

tepid trench
tulip sedge
clever briar
#

What really should be coming next is making the Covers feature free to everyone. Seriously, I feel like it should be out by now since it kinda is like Audio Input where you input a sound and genrate a song out of it.

ornate trout
#

I wish the replace section feature allowed me to specify a range length so that I could have more control over how much audio is inserted

tidal zinc
#

"Exclude Styles" does not work

#

I'm getting so tired of generating noises that I'm about to create an AI and train it to avoid the damn noises.

zinc marsh
#

So I have couple problems with suno I really hope 4.0 version will fix it , I been using pro version for about a year now , I started AI music group it’s called LAFA it has 3 different people , girl and 2 guys , so I guess u all know how hard to make suno do 3 different voices and how much credits it costs + I’m looking for most realistic voices so it annoying , so I really wanna gove couple suggestions. 1 save vocals , literally make feature that if u get really good voice you will be able to save it and use it in another song , it will make easier for those who making music for AI singer or band.

#

Will be awesome that u could choose those saved voices to sing any part of the song u wanna . Because right now you get perfect results on 1 verse now cords coming in you want to switch vocals from men to woman and at this point it’s just sooo frustrating…

bright fjord
#

there should be a place to choose male vocal ou female vocal. For we can decide. Because sometimes we want only women, and male vocal keep coming

jolly inlet
#

Wish the text in the video had a black border so we don't get issues like this

image.png
spice fog
#

Hello Suno Team,

Here are some suggestions for Suno:

  • Karaoke game app
    (because people enjoy interaction).

  • Fade-out / Fade in function for music/songs.
    (Because this is the easiest way to make people feel that it is a complete song, and the easiest to work.)

  • Keyword search for playlists / option to add playlists.
    (Because searching for songs always look down a lot of time.)

  • Invite a member to receive a reward of 10 days of subscription instead of 250 free credits.
    (Because songs that are rewarded without copyright are meaningless.)

  • Random play for music in playlists.
    (Randomly playing music will psychologically reduce feelings of boredom.)

https://suno.com/@tonkabean

Ai Bean Music | Join me on Suno

https://www.tonkabean.me/

https://www.youtube.com/@AI-Bean-Music/playlists

clever harness
#

Consider tweaking the colors on replacement

I finally got the hang of extensions today, and I think it's great. However, part of my learning curve was not really understanding what the current UI was showing me because color change was so subtle (gray to slightly darker gray). I think this could be enhanced by making the color of the replacement options match the color of the edit UI.

Now it's much more obvious to me that that fuchsia section above is represented by the fuchsia color in the replacement. It was particularly confusing because the replacement had some lyrics, and then outside of the section duplicated the lyrics, so it wasn't clear to me if what I was hearing was my mistake or Suno's mistake.

image.png
clever harness
#

Playlist likes on Profile aren't consistent with song likes

I'm not sure why this is, but unlike the songs on my profile which change the thumbs up color if I've liked them, the playlist thumbs up stays the same color. Why do I want to thumbs up my playlists? I don't have a good reason, but I do wish the UI would present this action consistently. 😄

Also unlike songs, when I click like, it takes me into the playlist. When I click like on a song, it doesn't take me into the song. I find this difference confusing. Ayyyye, while I've got you... you know what else would be great? Showing me the total run time on a playlist in the same way you display runtimes for songs. That would make for a convenient way to determine if a playlist is sized appropriately for an album/walk/etc.

manic cypress
#

We should be able to upload longer than 2 minute audio file and then Suno can automatically truncate into 2 minute long for our convenience. Thanks.

ember mountain
#

Other paymethods would be a banger

sage otter
#

more dynamic filtering options of user songs
eg. NOT x
y AND z
a OR b

modern summit
#

can you just like allow user to manualy adjust the timestamps on the replace section? i mean, why are they here but apparently not usable? it would be way more easier for people with screen readers and keyboard usage only

proper shell
#

maybe it would be possible to add the option of specifying 5 favorite bands in the profile, which Suno would later take into account when creating songs? For example, if someone adds "Linkin Park", SUNO will take it into account and create music based on this band.

patent nacelle
#

Replace Section should be available for any song, not just the ones under 4 minutes 😉

clever harness
#

My simple Lizard Brain finds this Confusing

Ok, so when I look at this my brain says, "play is active, thumbs up is active, and thumbs down is inactive" My next thought was, "I like this song let me click the active thumbs up button. I know white means active because it matches the color of the play button which is active for me to click."

However, that's not what this is ackchyually communicating. This UI is using one color to communicate two distinct unrelated messages:

  1. White means active
  2. White means you've already taken action

When I click on thumbs up, I'm actually removing the thumbs up from a @sage mirage track. On plus side, when I click like again his heart will be filled with joy when he gets the notification. What I'm saying is, I'd love to see a third color here to make this much more clear. In my own completely above the board Suno side project, I render Suno colors in: white, hot pink and deep sky blue. And let me tell you, like my collar on the dance floor, these colors pop.

Screenshot_2024-10-15_at_6.40.24_AM.png
grim badger
#

In the trash bin it would be nice to have a checkbox to delete multiple songs permanently

clever harness
inner forge
#

For cropping, it would be better, if we were selecting what we want CUT from the songs, and if we were able to select multiple segments at once, to be able to test/preview the seams, of the end result.

tawny pilot
#

i know this has been mentioned before, just putting in my suggestion for more voice control options, mainly using the same voice for multiple songs, and it would also be nice to control multiple voices in one song and what parts of the song are sang by each voice

icy valley
exotic idol
#

/chirp

rare obsidianBOT
#
The /chirp command is no longer working

Chirp is retired, and no longer on discord. You can make music now on our website: https://suno.com

lunar nexus
#

Hi, I love Suno, but sometimes we can create something really great and still make mistakes in the lyrics. Please add a feature to 'remake' music with the same instruments and voice, but with corrected lyrics ❤️ ❤️ ❤️

analog arrow
#

I am a SUNO subscriber, and I've tried the editing features. From the demonstrations on YouTube, I saw that it's possible to replace a section of vocals or add an instrumental solo. However, I wanted to add lyrics and vocals to a specific part of the music, but I couldn't find a way to do that. I also tested the crop function, but I couldn’t figure out how to remove a section from the middle of the track and reconnect the remaining parts to form a complete song.

I really enjoy using SUNO. It allows someone like me, without composing skills, to create my own songs. I hope more flexible features will be introduced in the future to help make the creative process even more perfect.

leaden marlin
#

It would be sick if you could upload a file, extend, upload a new file to influence the product. I want to hear a reflection of what all my original files sound like.

untold flume
#

library sorted by date please!

heady iron
# lunar nexus Hi, I love Suno, but sometimes we can create something really great and still ma...

I second this!

I have used InVideo and am able to work with their system to replace certain parts of audio or video, fix narration, change visuals at certain points etc without redoing the entire project and losing the essence of a creation, and it allows for pretty detailed modifications.

I just started using the trial of Suno and it rendered an absolutely incredible song but I was super bummed to hear I couldn't modify a few words of the lyrics, at the very least, and even more so, that I can't retroactively edit any songs that were pre-Pro version. That would be a great addition too.

I tried Donna too and it has instrumental and vocal stem splits built in which is nice, but Suno allows for so much more creative potential in the prompt features you allow.

If there is a way to save a Seed of the instrumental and Seed of the vocal and then make minor touch ups here and there it would be amazing. Either in the way the AI may pronounce a word, or a flow of a particular phrase, or at least just changing a couple words here and there, or a typo, and allow a re-render of the vocals over the track without losing the unique vocal style that was attached to it.

Hope that can come to life someday!

I'm sure you're all dedicated to leveling Suno up and I'm incredibly inspired by what you've already accomplished and provided us!

✌️😎

lunar nexus
clever harness
#

Include Artist image at end of video

One miss of the video feature is the end you should display the creator's image and their suno url, or even a QR code to take them to the page. It's great that Suno watermark is there, but I feel that the creator should also be highlighted at the end. 🙂

cosmic island
#

I would love to be able to select the instrumentation, and direct how that instrumentation goes eg, guitar, dark, heavy riffs, etc, right now the tags only feel like 'suggestions' to the AI, and having more control over that i think would add a depth in the instrumentation, I find that when it picks a distorted guitar for instance it wont change to a clean guitar in another section which makes the over all song flat and on dimensional.

heady iron
# lunar nexus I'm glad to see that I'm not the only one to need this ! Come on Suno make a v4 ...

It would be great to also have a track style and vocal reference template option based specifically upon a previous piece you rendered that you like most, and you can directly reference your next song creation on that seed/template. You could choose instrumental, vox, or both to be reproduced so you could create a more consistent style once you find a sound you love. I'm new to it and it's amazing what variations can be done but it'd be great to have more consistency if desired.

Could even develop it to the level of having presets to apply for vox, instruments etc. Lotta possibiltiies

viral ember
#

Trending Global Revisited - I am taking into consideration performance / load concerns: But have a few options to improve it: 1) have a larger Invisible Bucket - I think it shows 35ish tracks today (in a random order - this was a great change as only top X got any decent views before, thanks again for doing this!) - To improve further and maintain performance, perhaps have it draw the 35 visible tracks (decided on page load) from a larger bucket of between 50-100 tracks.

#
  1. This suggestion is probably harder to code: have a certain number of the track slots dedicated to rotating around the top tracks from the most active Language pools - right now English numbers brutally dominate the game, which doesn't help promote Suno as well as it should in other countries, regional charts help, but can we go further? maybe Dutch, Italian, German ETC. should always have a minimum of 1 or 2 slots on Global Trending?
#
  1. Spread the love - Cap the number of songs PER artist (and PLAYLISTS) to 1 per user at a time - THIS IS THE MOST IMPORTANT SUGGESTION HERE TBH! It also stops malicious users from spamming to fill up the trending / playlists, especially when contests are on too ensure NO ONE ELSE gets likes / heard. (we have seen this attempted!) - Initially a simple rule like your NEWEST public song takes preference if eligible to also be on the playlist. - later on maybe the super user could have a page where they can see all 'trending eligible songs' by region and choose which one they want to actually be public for that time window. -- for snapshot of data I think I think 7 / 35 slots (20% are people with multiple tracks) - typically we always have people (or multiple actually) with 2 or 3 tracks on trending.
opal spoke
#

Extratone genre/splittercore

true coral
#

It would be nice if there were a check box when you extended from uploaded music that you don't want to necessarily extend from that music, but to generate the new song using melodic cues from the uploaded piece.

grave fiber
#

Here is my take on Suno. Honestly, one of the best tools I have ever used. The reason I say that is because I’ve been a songwriter since 2018. I used to write songs for singers to perform, as I hired them. As the list of songs grew, the singer, sadly, left to do his own thing, and I’m happy he did because it was something he really wanted to pursue.

With that being said, I started writing music with a backlog of maybe 90 songs, but only had 40 that were ever released. When I found Suno, I was shocked to hear that I could put out the songs with my own lyrics. So, after a while, I started using Suno to sing my music. But when the upload feature came into the picture, and I found out I could use my own voice, I began using that feature. And I’m so shocked by it. The songs sound like me! Wait, what? Me and my friend, Lovely Soul, wrote many songs, and now I can sing them because of the upload feature.

The only thing I want to say is, please make it so that when we put our voice into a song on Suno, we can make it public. I promise you, it’s my voice or Skippy The Evil Clown’s voice. I think I’m addicted to using my own voice, and I have a lot to be thankful for—Dreamer and Daizy, to name a few, and Lovely Soul, for pushing me to do my own singing. I want to thank you, and also Suno, for making something so special to all of us. I’ve even thought about, maybe one day, singing my songs live that I made on Suno, on stage.

tidal zinc
#

I suspect that Suno doesn't want artists to use the tool professionally, but rather that it be used only to pass the time on the site in an amateur way, generating anything and for the whole world to use, considering that they are having unnecessary updates like this last one that doesn't add anything, while we take it more seriously and seek to improve the ideas by working together to have more autonomy...

modest frigate
#

we need a "FIX" function to only fix the sound with a new identical generation and this should work also for whole 4 mins songs.

loud kite
#

Finally one of my suggestions gets created Suno scenes is basically the video to audio feature there's been a lot of papers on it but Suno lagged a bit on implementing the research into a product.

It would be cool if you could also do the reverse it makes a music video based on audio just need to use good annotations. Right now gotta use runway and slap a Suno track on it but Suno scenes could make the process more end to end and seamless.

royal lichen
#

Add feature to gift monthly subscriptions to friends. Life of quality update

humble drum
#

refund option for covers that are just the original spat back out. that's about 80% of them. 600+ credits wasted...

unkempt zealot
#

Multi-Select / Select All

Let me delete all or select multiple tracks at a time inside the trash bin for deletion.

Zooming for Replace and Trim

The ability to zoom in and out on audio to achieve more fine-tuned trimming /cutting.

Song Voice selection

Really wish we could pick the style of voice we wanted for our songs.

split fog
#

Add a feature where you can make parodies.

north fable
#

I would VERY much like the ability to download a zip file of a song that contains all of the various tracks within the song as individual, separate tracks.

That way I can import them into my other audio programs and tweak it to be JUST the way I want it.

jovial raptor
#

Automatic partial refund for songs you delete in certain time window (delete option to save the file for a certain time to prevent using it wrong). Right now most of the songs goes straight to trash.

hushed breach
#

Suno android app?

autumn mason
#

Please make vocal separation better. Sound like crap now. Since covers can make awesome re-creations with vocals. Not just like it sounds now AI separation. I mean dry vocals just the main lead. No harmonies, doubles, unison. Just main vocal clean. Thank you for an awesome product and tool as I see it. Just missing clean dry acapellas.

lean zodiac
#

Can you create anti-label research?
Like display all song not liked or all song not-published ?
For we do some cleanning in our songs created ?

lilac badger
#

How about something similar to the cover option instead allowing the ai to come up with a result using two songs as samples and then making a mashup?

lilac badger
unkempt zealot
untold flume
#

discord bot please

wispy frigate
#

Releasing new features to iOS users first makes me want to post #Udio tutorials.

modest frigate
#

In the meantime I use more Udio than Suno 🙂 Or both, it is called Sudio 🙂

potent tapir
#

I think having producer/composer friendly things would help us use it as a tool and not see it as a threat (I mean, in a chicken little says the sky is falling sense, I welcome our AI overlords). Get stems needs to get more than just vocals, so I don't have to keep using RiffusionPlayground's cutter which gets more [sounds] than FL Studios at least. Train a better end result for final song output when doing extentions or covers of my uploaded work, since a lot of my original stuff gets terribly mangled in the process, and I end up with terrible phasing or original parts that lose dynamics (such as fast, choppy, stutters and arpeggios).

potent tapir
lilac badger
modest frigate
#

sound quality of Udio is higher than Suno's. But the default settings are wrong, and it takes a while to get the right ones.

clever harness
#

Option 1: Personalized Global Trending

I have nothing against the current trending superstars, really, honestly, I mean that from my heart. What I think would be much healthier for the community is personalized global trending. Show me the most popular music based on your understanding of the music that I enjoy based on the songs I write by people I don't currently follow.

If I've written no songs at all, or you don't understand what I like, sure fall back to global trending. Over time this eliminates the importance of trending as a sort of Suno meta game that people play, and ensures I'm only ever shown fresh faces that I might want to follow.

Allow me to downvote the songs from the context of trending. This would mean: "I don't want to hear from this artist," after x many votes, and so I'd never see them on trending. However, don't penalize the song writer, it's not their fault their style doesn't appeal to me.

clever harness
#

Option 2: Proportional Trending

Look, if you have a ton of fans, that's great. It's no one's fault but my own that I have no friends and write mediocre music. That said, imagine you join Suno as a fresh face and write fantastic music. The current system is demoralizing to those people. I'm comfortable being a friendless trashcan man, but I empathize with those that lack my self confidence.

If the trending score (however it's determined) was proportional to your fan base, such that as your gained more fans you ability to trend would become harder and harder, eventually you'd stop trending. This would allow new people to come and trend. If you were doing truly fantastic music, then great build your fan base some other way, clearly you know what you're doing.

You could of course keep opening unlimited free accounts which of course be a ToS violation so you should get tossed, or you have to spend unlimited dollars signing up for paid accounts which only helps to support Suno.

clever harness
#

Option 3: challenge-response authentication mechanism as a solution to trending

Now I know there are not bots or people manipulating trending/contests on Suno. That said, imagine some world where maybe in the future, someone did try to manipulate trending. I know this won't happen, but let's go through a thought experiment.

What if you could solve any suspected bot/bad boi behavioral problems related to trending by only considering likes given after some arbitrary period of time unique per play session and user. How do you prevent faking this out? At irregular intervals the client must ACK back a unique time-based token provided by the server.

Basically using a challenge-response authentication mechanism to ensure plays check-ins are legit. However, only one of these ACK is the milestone marker that means a like past this point in time is counted as a trendable like

The client has no idea did their like even count as a trendable like all they know is they liked the song. The artist sees a like and they are happy as well. This increases the difficultly of gaming the system, because the bot can't solve the problem because the solution is random.

You'd need a bot or a person to listen to a song in it's entirety to ensure the like would be counted as trendable like which increases the effort of gaming the system. You might even just throw likes that came in at the end as counting towards trendable like to avoid the problem completely.

It also means trending is no longer a pure raw popularity contest, you could actually get songs with less likes that trend. There are of course challenges of knowing the appropriate start and end range for this kind of threshold, but I assume Suno has plenty of data

spice fog
wicked galleon
#

i need suno v4

celest coyote
#

Make an Android app

tender saffron
#

If there is one? I cant find it. If I have trash And I have alot, instead of deleting 1 by 1 seems like a 3 step process. Have 1 button delete all trash. If there's a post on this already i apologize i did not see it.

#

I do apologize there is quite a lot of post's on this issue.

hidden furnace
#

Suggestion: allow playlists to loop (first song plays again after last song ends)

gusty ocean
#

How is it STILL not possible to download all songs in a chosen format?? Amazing to me that this simple feature still doesn't exist and we have to manually download one by one.

silent cipher
#

My OWN Voice profile:

Anybody thought about a feature where user can tape his speaking or singing voice in order to profile and exchange with the AI voice now? that would be a great feature, and it should be in a way where the own voice can be add-on on to the songs later.
In general, changing the stile of voice after the whole sound is recordet would be one of the most required features in my view!

zenith rivet
#

It would be great if we could sort playlists after view count, like count and newest or shuffle them, besides the one the playlist creator choose.
This coul help a lot 🙏

gloomy hound
#

Idea:
Make regenerate video available for all users (with a daily limit maybe)
And give Pro users the possibility to regenerate without watermark (since they own the song anyway)

This would be fair for all users i think!

raw mist
#

Please work on getting clear clean vocals. Currently they are distorted and non-professional.
Thank you.

tidal zinc
#

next update: another silly thing for amateurs to play with

glad sparrow
#

Please in the pro version, be able to choose the creation with the vocal already divided without shadow and without needing to do the stems after as it loses quality.

tidal zinc
#

We were all hopeful and eager for Suno to evolve musically, but it seems like they just want to expand their tool to get new users, we don't need notifications of likes on the site, we don't need to create a video before mixing and mastering the song, we just need to make our songs

tidal zinc
#

2,500 credits, I made two songs.. 😂

glad sparrow
wanton rover
#

As a paid member not feeling Suno Scenes they need to find a way to keep the same voice going if one is trying to make a album using that same voice that would be better than Suno scenes this has been a ongoing issue for 3 months now

#

The other issue is having the same voice with a different instrumental

#

Instead of burning through credits

jovial raptor
#

Not sure if anyone actually asked for scenes—maybe it's just a way to appeal to a wider audience. I've seen many software projects explore other areas and lose focus on their core purpose. I hope that's not what's happening here.

clever briar
velvet horizon
chilly bluff
#

i have a melody which Suno produced from my lyrics and a prompt. I really like this set of features, music rhythm, voice characteristics, etc. is there a way, to get those characteristics from Chirp, so i can use them in a new version of the song?

frosty breach
#

Gifting/Tipping (paid) creds to the creators you particularly enjoy. would be a neat function.

cursive crow
#

I would love to be able to use my own voice. Others have said the same thing and it really would be amazing.

hot pier
#

would it be possible, as proposed by MOISE AI that I also use, to have the chords of the songs (and to be able to download them) and the overall tonality of the piece. in the same vain request could we have the possibility of choosing the tonality for creating a song

clever harness
#

Automatically regenerate video for song when lyrics are changed

I like lyric videos. I like Suno mobile app. I like suno.com. That said, what I find super confusing is
when I update my lyrics on suno.com it doesn't also automatically regenerate the video. Why is this important? It's important because on the iOS app, the lyric display is driven by the video, so all my songs are out of sync unless I remember to manually go through this step.

I can understand the concern that you don't want to load up a queue with a million requests to generate videos for the same song, but you could operate a queue off the song id with some window in time for generation. For example perhaps you have an n minute window that must passed prior to a regen a video on save such that 1,000 edits within this time box would still only result in 1 regen.

"wow, Alex that's a great idea" whispered behind my own hand.

What who said that?

modest frigate
#

<@&1108435550475653161> Maybe you think about adding another plan. 10000 is a lot for most people and 2500 too little.
Maybe you should add also the option for 5000 credits. Since the people can use 2 different accounts to achieve 5000 credits anyway.
So why don't you offer this directly ?

sturdy wave
#

Would like a Shuffle/Random option for Playlists

sage mirage
glossy ginkgo
#

Add Discord to the "Share to" options on the website

hearty ivy
mint burrow
#

The image upload box still crops to a square but images are now shown in portrait! Please update the upload box crop to show what the image is truly going to look like. wink 😉

acoustic creek
#

Will there be an option sometime in the future to search uploaded songs in the library?

zealous ermine
#

Hello, I'm a screen-reader user and would like to suggest the ability to use the keyboard to input the timestamps for where I'd like to crop or replace sections of my song. I was trying to do so (as it seemed to allow me to type) but whenever I do, it clears and I cannot perform this operation. Thank you very much.

scenic patrol
#

Has there been talks at all to select specific parts of a song to add in specific instruments or elements?

inner forge
#

Can we PLEASE for the love of ANYTHING, have a toggle, as to whether we want Suno to get creative with our lyrics, or strictly adhere to them PLEASE!

Other than that, right now, the only suggestion I have, is to get Suno working. I have wasted two months of credits, and upgraded to premier, after I blew all of my pro credits, just to be able to afford to retry ONE song, ONE... that Suno can not get right, no matter what I do. I am trying to cut, and extend, and replace, and start over, I even wrote new verses, thinking maybe I needed to fill some sort of gap, and Suno just continues to get the lyrics wrong. It favors one verse, in particular, and repeats it when/wherever it wants. This was supposed to go on my Halloween album, and I feel like it's just impossible.

I appreciate Suno, and what I have been able to do with it. I want to see Suno succeed, but, knowing what it can do, makes all of my struggles seem very unnecessary.

patent nacelle
tranquil sluice
tidal zinc
clever harness
#

If I do a "proper" lyric video, I just make my own and don't rely on the Suno ones. 🙂

violet sierra
#

Would still love a low, medium and high cover strength option.

#

Right now it can be any of those randomly.

lilac carbon
#

1 - It would be great to see a highlight on the lyrics of the song while the audio plays (like a karaoke).
2 - It would be great to be able to give separate ratings for the style created, the singer's voice, and musical arrangements, and based on that SUNO could create the next tracks with the same style.
3 - To be able to move a slider to get the percentage of influence of the Cover over the included style or vice-versa.

lilac carbon
humble grail
#

Bringing this msg from the channel #📚┃suno-school
[Covers Feedback]
I wish I could set the level of creativity when creating a new cover. Some of them look extremely similar to the original. At least when I'm trying to change a Trance Song to a Metal or Rock Song.

hearty copper
hearty copper
# lilac carbon SUNO is the creator, but it can't always do everything, it has to be edited, jus...

I think their point is more along the lines of "Suno added lyrics where I didn't put them," as opposed to "singer was out of tune or the instrumentalist can screw up," I too have seen this, Suno knows it has 4 minutes for the song and then at the 3:50 mark tries to throw in one final chorus that every other time has taken 20 seconds to complete so you get (using chorus to Islands in the Stream) Islands in the Stream that is what we are/Sail away with song cuts out with no fade But yes, like you I too am fairly happy with what Suno does

#

1 suggestion for wording 1 suggestion for the system Suggestion for wording: if the browser can do something, maybe not say it is only available in the iOS app (The browser what's new has a link to use Scenes) Suggestion for the system: Some of the songs are awesome but cut out abruptly, maybe have an editing command that allows you to add a fade out, yes you can use an audio editor, but some of us don't have them

spark flame
#

Allow us to download seperate audios. Vocals only, SoundTrack only, or both combined

hearty copper
rancid pasture
#

Ability to alter the apparent skill or technical skill of the singer would be really cool. Getting more authentic GarageBand sound.

zenith rivet
#

If you upload an audio with vocals it will write them in the lyrics.
But for some reason this detection system is seperate from the one that blocks it from setting these songs to public.

So my suggestion is that it should be connected and only when it detects vocals with text it should be blocked to set public.

It has some use cases, for example false positives or animal sounds!

zealous ermine
# tranquil sluice same problem. scereen reader user too.

Glad I'm not the only one. This (being able to use the keyboard to indicate the timestamps for what we'd like to crop or replace), would be incredibly useful. I have been using Suno for my podcast's music, and it's been a great pleasure to integrate it into my workflow.

tranquil sluice
frosty breach
#

Hey, there seems to a conflict with bulk removal of gens. It seems like if the types don't align it screws with the ability to delete. Also some full songs, depending on if you're in create/library, are unable to be removed as well. Similar issues seem to happen between different part types, like part 2 and part 3s in bulk.

#

wrong chat but I don't wanna DL the recording.

true heart
#

don't know if this is already possible, but would like to suggest an expansion to Covers, where if you upload your own vocals, have suno gen the backing track, then I'd like to be able to remove the vocal, so that you get a karaoke backing track

quartz lagoon
#

It just makes it feel more authentic

nova flicker
oak sapphire
#

HI @strong swan Mr Tree - why does the REGEN VIDEO not change the LINK LYRICS to match the new lyrics sync (in the LINK version of the song, Yellow Circle), but the LINKED LYRICS seems to make it into the DOWNLOAD section... is this by design?

Can we get a NEW VERSION of the video generated, via CREATE (instead of just before download), that might cost credits or not, and have it generate a new version of the Replaced Video with the new link?

This is what I actually was hoping for.

Suno-Web_Regen-Video_vs_Link_20241018.png
strong swan
royal girder
#

When adding a song to a playlist, prevent adding it to a playlist already added, maybe gray out the ones that its already been added it. I typically end up adding the same song to multiple playlists over time and sometimes ill end up having it added twice, not songs of the same name but songs that are the same, luckily removing one doesnt remove both. So its manageable

umbral mist
#

And as far as suggestions, 120 character limit for prompts is wayyyy too small of a field

spice fog
#

It is hoped that the playlist can have a selection mode for display, a text-only list or pictures and text.

If it is a pure text list, it can be arranged according to update time, or A-Z.

true heart
leaden cradle
#

I would personally love to see more effects added as features to the already implemented 'crop' feature, such as 'fade out' and 'fade in'. A lot of tracks often end while the song is still ongoing or abruptly starts in weird positions.

carmine solstice
#

It would be cool if we could change the language of the song like from English to German or Japanese to Spanish just by using AI.

royal venture
#

Yall need to do better job at protecting users music. Not cool seeing my song stolen, uploaded in China, and the song does better than anything I've ever made here. https://i.imgur.com/Wr0kIhx.png

Imgur
versed sandal
#

I may have suggested this before, but I'd really like to see a toggle that lets you create a looping track, for if you want a background track for a video game, "Please wait" screen on a Twitch stream, etc.

royal venture
#

Who ever just said that is an idiot. People shouldn't be able to download songs they didn't make. They shouldn't be able to look at html code and have download for everything just there for all to find. Btw I dont care if they take my song and upload it to some random site thats unable to monetize it. No one can upload any of my songs anywhere that matters. Its still annoying to see. They could be listening to these stolen songs at the source instead. You should read TOS which says you give permission to people to use your song when they have the audio file generated by suno when you download. Since people can obtain this without permission, even if the song is private, they should change that. Especially when that file is downloaded and reuploaded elsewhere for anyone else to redownload. Sites like the one i showed above have literally every song every user has made, albeit on a delay. This is ignoring covers which are non-consensual so the user can say they made the song which they didn't.

lucid gate
#

scenes on browser

#

please

rotund maple
#

I just want upload audio to be longer than 1 minute

blissful sail
#

Looks like people are able to e.g make a cover of your song and then publish it with zero references to the original. Would be great if the original song was at least visible somehow as well as the original creator. I don’t mind this as long as I’m credited.

clever harness
twin hinge
#

Suggestion: Allow us to revert back to our original lyrics in the song details. ♻️
There are some lyrics I deleted not knowing I would not be able to get them back.
Additionally, with the new replace section feature, if you deleted your lyrics and forgot to save them, you cannot use this feature at all.

twin hinge
sterile stream
#

Suggestions:
Yearly Suno awards
Suno Radio
Ability to co-create songs (with all profiles being credited)

#

💕

nova flicker
nova flicker
random remnant
#

I would like to see folder hierarchy added to the browser interface so that I can better sort and order/clean up my 1000s of songs. It would help with sorting keep/cut decisions on song renders, etc. Thanks

stone steppe
#

Hi Suno.

Please upgrade "replace sections".

Thank you

naive solstice
#

Add Folders in "Create"

nova flicker
#

Mainly Web Recommendations

  • Add to multiple playlists on selection

  • Confirmation on move to delete

  • Title and Genre tags not links in main list, just sidebar

  • Title is editable from selecting in main list

  • Exclude styles colored differently

  • Sidebar genre and styles have an extra space (but shouldn't)

glad steppe
#

wished we could select generated cover arts from previous generations … too many times the generated cover art fits but the song itself got messed up.

wooden fable
#

Sorry for sounding like a broken record, but could we have v3alpha in the model selection drop-down. It's the only model i really liked. It's the reason I paid for a year. I find it far superior to v2, v3 and v3.5.

feral aspen
# wooden fable Sorry for sounding like a broken record, but could we have **v3alpha** in the mo...

I too miss some unique aspects and dynamics in the alpha that I could never quite match in later versions. But I have some good news for you. I think you may pleasantly surprised by the new Cover feature.

Instead of thinking about it as a traditional cover of an already complete song, think about the cover feature as a second rendering step in the production process after generating a starting point with V3.5. In fact try to think about all the new features in this way. Replace is another new feature. You could use for fixing a typo in your lyrics. But Replace is also a feature you could use to dramatically transform a song, not as a corrective feature, but as a generative feature.

With the combination of these features, I think you will find you can get to many of the same places that you could only get to before in the alpha. Maybe not every single one, but on the other hand, you can also get to many new places the alpha could not reach.

wooden fable
#

Can it remove the reverb from the vocals?

#

i mean, those are cool features, but not really what i'm looking for tbh

feral aspen
wooden fable
#

oh!

#

i'll try that.

feral aspen
#

I'm slowly coming back online from some life stuff, but I'm sure I can find a good example of de-reverbing a song later.

wooden fable
#

for the back log completeness: i'm gonna experiment with the cover feature (already getting some results), but i still would love to see v3alpha back.

naive solstice
frigid shadow
feral aspen
frigid shadow
#

I was asking because some of my songs exceed 7 or 8 minutes, I'm pretty sure those can't be properly covered.
Checking right now

#

Yeah, it doesn't work

royal venture
# nova flicker Whether they should or shouldn't, for as long as people put things on the Intern...

Obfuscate the url or create a temporary download link for the song and not one for where the original is stored and prevent users other than creator from downloading the original file of a song, which allows anyone to claim ur song is theirs, or you gave them permission to use it. On top of covers, extends, downloads, remixes, etc having to be requested and approved. As in theres a button to request this. In notifications a user can approve these or deny these.

No other website has this info shown in html code, or anything. Suno is intentionally allowing theft of not only creators work, but their entire website. The funny part is their websites are better.

Who wants current song page over this....

royal venture
nova flicker
royal venture
# nova flicker my brother in christ, it is not difficult to download something; sure they could...

Try downloading something from apple, spotify, etc without paying. Try downloading any paid software. Yeah you can torrent, or record the song as it plays, which is already going to be garbage, and then upload it somewhere and the audio will get even worse.

If its so easy everyone would have the best software for free, any song they want for free from anyone, any movie for free. The fact is it doesn't happen, and torrents are a dying relic of the past filled with nothing by trojans and viruses idiots download along with what ever they wanted because their paying with all the personal info and data.

nova flicker
royal venture
# nova flicker 💀 it is that easy, it does happen, you're just ignorant

Ive been on the internet longer than u been alive. I look at website code all the time. Its not easy. Only way you get anything is if someone pays, and then gives it to you somewhere else, or they crack it, and then send it to you. You will never be able to just go to the source and steal anything.

nova flicker
#

lmao

#

you're making ai music publically, you're not paywalling and your boast of possible age or experience is actually very laughable, so thank you for that

jagged trail
#

Any chance you can give us the ability to buy people points, sometimes i go to someones page, and I feel like I wanna buy them some points to keep them making stuff.

#

You get some more of my money, and they get some points, this is a clear win win.

royal venture
mild cairn
#

Feature Suggestion: It would be pretty epic, if we could (just like how we can Get Stems) kind of use that to generate a pure Instumentals track, and I mean like a completely new pure generation. Currently it sounds like it's just isolating stuff - if we can get a pure instrumentals, heck maybe even a pure vocal, from a completed song, that would be so epic. By like, using the stemns to create high quality new generations that are literally exactly the same. Just regenerated to get rid of all the noise. Make it cost 20cred, IDGAF, but the extra quality would be sick. And help a ton when mastering.

tepid trench
#

@nova flicker @royal venture This channel is used by the devs to check for feature suggestions. You might disagree with someone else's opinion but this isn't the place to argue with others about how useful or valid a suggestion is, nor to resort to exchanging GIFs (which have now been removed.)

vital jewel
#

the ability to upload 3 minute audios

lilac carbon
keen burrow
#

👀 Could someone please tell someone else--to please make the browser NOT auto bump down to the bottom of the page whenever a song has finished generating? 🙄🙃

leaden marlin
#

Hello may I ask if is it possible for the cover feature to be access with free users?

twin bear
#

Suno website, and mobile app, should allow users to not only follow each other, but allow for messaging each other too, group messages, etc. To fill in the blanks and give Suno a social media feel.

vital karma
#

Would love if people were able to add a language tag to their song, making it possible to search for songs with lyrics in your language.

clever briar
queen galleon
#

I'm sure these ideas have been mentioned before, but I have a few thoughts that may benefit us all.

Having the ability to reference another song on suno while creating a new song in order to get results that have a similar vibe/sound/genre/etc... would be incredibly useful/helpful when trying to make an album of music. Sometimes Ive spent 1000 - 2000 credits just to find a song with a similar sound to what my album is meant to sound like... (I still have uncompleted albums due to this issue), but if I could reference my other song in that album to have some consistency for my album it would be one of the most useful features on suno.

In addition if we were to have the ability to set a key, key change, duet, harmonies etc... that would take the music to a whole other level. Sometimes I get lucky for duets and random key changes, but it is rare and this really should be a prompt within the music like [Key Change Up 1], [Key Change Down 3], [Male and Female Singing Duet], [Two Males Singing Duet], [Male and Female Harmony], [Two Females Singing Harmony], etc...

Speaking of various voice styles, perhaps having a section in the music creation that indicates if the voice is male, female, a duet or a group singing the song might be useful. Just a thought.

Lastly, expanding on the cover and section replacement features to be for any song length and having the ability to expand/extend songs to the left (creating a new intro for songs that didn't generate one) would be amazing as well.

Adding these features/prompts to suno would be a game changer for our music and provide better song creation and control by the artists in general. So I hope the suno devs will take these concepts into consideration. Thanks!

frosty breach
#

The ability much like Apple to colour code your generations. Here's an example. That way I can go through all my generations, give each a little colour that I deem to be kept around for further 'Sunoing'. Also makes it a whole lot easier to not accidentally delete anything you didn't want to be deleted. I am currently using Likes for this, but it prevents me from liking other people's stuff as to not ruin my bootleg sorting system.

Screenshot_2024-10-20_at_16.27.42.png
lucid gate
#

add scenes on browser now

twin grove
#

I'd like the ability to mix Cover and Replace Section, if that would somehow make it so you can more easily surgically replace lyrics in your existing melody

frosty breach
#

Darn, I guess it was you then.

tepid trench
#

Woodhouse tells me this is what you wanted to say.

Screenshot_20241020-1651302.png
tepid meadow
#

when suno 4 comes out try to make it so that the instruments sound way more realistic then they alr are 🙏

outer berry
#

Devs I'm gonna need you to turn the fuzzy search on the search section back to a more literal search. Just how it was weeks before. I can't find recent songs in my native language anymore.
Thank you beforehand for your work.

#

Also playlist are cool but what I need is the ability to create BILLBOARDS. That update themselves on a function of time and likes.

inland silo
#

maybe it is a idea to add a option to only generate acapella's.

sick lion
#

Wish Suno would tell you if a song belongs to a playlist - either the creators, or if it has been added to another users.

fluid cedar
#

What if suno added a mashup feature, that way you can combine 2 songs into 1 song (just my opinion)

amber finch
#

Multi-lyrics-boxes for multiple singers would be cool

prime rock
#

Can someone please explain why the “make same length as selection” button does nothing when replacing a section? It just gave me 2:10 of someone literally singing “la la la la” when I tried to replace a ten second section. So I flipped the button thinking it might be reversed and it gave me random lengths. Surely either removing the button or making it work should be a priority?

prisma aurora
#

Please let me upload my vocal clone to use as the voice for my songs 🙏 🙏

twin hinge
upper basin
#

I wanted to propose an idea for expanding Suno's capabilities and shaking things up in a big way. Imagine allowing artists to upload their stems from their DAW, and your AI could analyze and replace their sounds with higher-quality, sonically pleasing alternatives. Then it could automatically mix and master the stems to a broadcast-quality result. Exactly like recording at a major studio with top engineers. This would revolutionize everything! There are still people who highly value creating from scratch, but when it comes to sound choice or mixing, their tracks might be falling flat, but if you created a very intuitive sound replacer and mixing tool, there'd be nothing stopping Suno from cutting a MASSIVE slice out of the recording space!

teal mountain
#

I have an idea that I think would be handy to all of at one time or another, it would be like the "crop" tool, but to make a nice fade out, maybe the highlighted part would be where you want the fade to start, you could make one for "fade in" as well.

potent tapir
# upper basin I wanted to propose an idea for expanding Suno's capabilities and shaking things...

I love it when it basically does a cover of my work, and it's practically the same song I uploaded but remastered and sounds like someone found the originals in high quality... sometimes.

Other times it's frustrating and doesn't make any changes to the song at all save for maybe an insignificant portion or second, when I'd like it to help me expand upon an idea of my own. So I think better control of the weights and parameters going into the model are needed to make functionality like this a thing.

Would be nice to have more control over things but I'm sure that's already been suggested, but I would think a fine point sticker needs to be put on it's name tag. Suno does much better at giving me the results for my own production needs vs. Udio, so I'm sticking with it for now, but it really could outshine Udio if it was geared toward producers as well as hobbyists.

frosty breach
#

Searchbar in trash would be neat.

vagrant orchid
#

SUNO! Instead of doing giveaways, spend the money on training compute to improve the model.

chilly stream
#

!SUNO I need see in the library list, Playlist label to know is asigned the album

prime rock
upper basin
# potent tapir I love it when it basically does a cover of my work, and it's practically the sa...

Suno should drop its own AI-powered DAW—imagine this: whenever a section isn’t hitting right, you just tell Suno to swap it out with something better—whether it’s a new idea, sound, or design. You’re still in control of the arrangement, but the AI steps in to level up the sound and mix. And when you’re ready, Suno’s DAW gives your track that polished, broadcast-ready finish using AI magic.

echo surge
#

🔥 FEATURE REQUEST: The AI voices in Suno are amazing. I would pay if there was the ability to upload a melody (and lyrics), and have an isolated Suno vocal returned.

⭐ Often Suno generates a great-sounding vocal, but the words or phrasing need tweaking. Ideally I'd like to be able to take the vocal stem into a program like Melodyne to get the pitch changes and phrasing correct, then feed it to Suno with the correct lyrics and select a voice (or opt to have it use the same voice as before). I would literally pay an ongoing subscription for this and every time I mention it to other musicians they say the same thing.

potent tapir
spring marsh
#

An Upload Audio filter please alongside the ones already for Scenes /Liked / Public so I can easily which tracks are my uploads 🙏🏾

#

In the Library

tulip sedge
#

FOLDERS! Please 🙂

clever briar
#

Please make the covers feature free/available to all users. You will recieve greater feedback on the service by including those who can't afford a subscription.

uncut otter
#

Library management
filter by

  • liked
  • disliked
  • unliked

Select all -> move to trash

I'll keep saying it, your model is great but your library management is dogwater

On streaming services, you are the one who adds stuff to your library. Here, you "add" everything you generate, even if 90% of your generations are bad and you want to get rid of them.

With no genre sorting either. It's near unusable.

#

You keep adding random features such as "suno scenes" which add bloat instead of improving your core product, and i get it, you're trying to scale right now and retention isnt' your main concern. But this stuff is rage inducing if anyone wants to use your product at the levels you offer (2500/10000 credits a month)

modest frigate
#

It's been exactly 90 day today (3 full months) since Suno HAVE NOT DONE any change, improvement, bugfix in the user interface or the song management.
This must be a world record, specially for a company with a web based software used by a lot of users (thousands, millions ?).
Congratulations Suno! And there is much to do 😦

image.png
noble osprey
#

Have a feature where instead of isolating the music with AI you have the AI generate a new isolated track. Stems is cool but it doesn't sound that good. Have the AI anaylize the song and create stems and allow us to choose what we want to keep and remove for a new isolated track. I want to dub over a ton of songs and I wont until I can do this

icy vortex
opal spoke
#

Proposal: Vocal-to-Instrument Music Conversion System

Objective: This concept revolves around a system that converts melodic vocal performances into instrumental compositions. The system captures a user’s vocal rendition, analyzing musical elements such as pitch, rhythm, dynamics, and melodic nuances. It then translates these patterns into notes and chords, reproducing them through virtual instruments tailored to the characteristics of the melody or based on user preferences.

ashen pilot
#

In suno, there is the replace section. Is it possible to upload another audio to replace the first audio section with the second?

placid garden
#

Big ask #1: Ability to output a file (even just .txt) of all my output.
Include song names, prompts, lyrics.
Big ask #2: Assign any track to a song in development and to an album. This is essentially folders. A plus would be the ability to export the track, album and artist as metadata for paying clients. Lots of us have multiple albums worth of material, and numerous tracks of a single song.

olive scarab
#

It would be nice to be able to record long prompts with arbitrary word prompts. I would like the ability to create and edit a personal library like a function. Also, this library should be able to be made public or private. Private if you want originality. If you are philanthropic, you will choose to make it public. Perhaps it might be worth the price. For example, I think you would want to try a prompt made by a famous musician.

lofty bear
#

As i just found out suno doesn't keep all my previous songs for obvious space saving, i want to suggest a feature for the website to store a small text file per user with just the generation data. Prompts, model if there's multiple, and seed that was used. That way we can not only recreate songs we've forgotten to download in the past/lost the songs, but also make other songs/similar songs on the same seed, just slightly different.

tepid trench
lofty bear
tepid trench
#

There's a page selector (top right)

image.png
#

as you can see, this is a song I have from back in February. It's on page 200 of my library

lofty bear
tepid trench
lofty bear
#

Checked via number and a few other gmail accounts, still nothing :S

tepid trench
stuck mountain
#

Can we download all the librairy in just one command ? In the format wav or mp3 ? It will be soooo cool

latent scarab
#

Please let me infinite scroll up in the create page on desktop, it's insane I can't go back to songs 15-20 away

burnt bridge
#

Please. Can we customise how many songs are on one page?

eternal pendant
#

maybe it so you can change our styles after making the songs

leaden marlin
#

will there be a feature where you can change lyrics?

frosty breach
#

00:00 extends Prayge

true heart
#

suggestion for the crop feature: let us crop a section in the middle and keep the stuff to the left and right - I can't be the only one that's absolutely hated a transition but loved the stuff it transitioned to.

oak sapphire
# strong swan the regenerated video is the one you share with the symbol in the yellow circle....

Thanks @strong swan for the swift reply... I tried the YELLOW CIRCLE share-a-link button (after selecting regen video under the download menu) and posted it twice in Discord adding "/" immediately and also with a space after the link. Neither worked. Clearing cache or using a different browser didn’t help. It seems broken.

Please let me know that you have added this to the myriad of suggestions on the fix list...

... if the Suno Team moves the "regen video" under the create menu, and it makes a new entry link that says (regen) after the title, and it actually works with the new link, this would make more sense. NO need for a conflicting web-cache record.

@inner forge or @still parcel - - Have either of you witnessed this (regenerated video link not working), or got it to work with the "/" trick ?

Suno-Web_Regen-Video_vs_Link_20241018.png
modest frigate
#

Is there reason for only max. 120 characters in the prompt? It is not really much and not enough either.

naive solstice
#

suggestion make donate button, where guys who like songs can buy credits for guys who make songs, and suno platform have extra money with that

true heart
#

suggestion: if you @ someone in the edit display lyrics, it automatically hyperlinks to their profile

tepid trench
#

The way the sub-menus open on the 3-dot menu constantly feels really awkward.
Because the 3 dots are to the right of the page, it means there's not enough room to display the sub-menu to the right, so instead all sub-menus open to the left instead.

image.png
brittle hatch
#

can't wait to test it , i wish there is away we can upload 2 audio one lyrics and one instruments that would be mind blowing !

open pine
#

Suno needs to add a tempo and pitch editing thing so you can get just the right vibe without having to reupload everything. Maybe you want to use a different genre than you want because it'll get closer to the sound you want but it's too fast or slow otherwise. Maybe you want to change the genre of a song in covers but your current tempo isn't fitting the genre you want to use. Maybe you just like what you have but it's just slightly off from your idea.

leaden cradle
#

Can we please get an option to mark all notifications as read?

compact hollow
#

Getting stems from an instrumental should give you the different instrument's stems instead of just splitting instrumental from vocals.

regal oyster
#

Hey, can you please add a function where you can select multiple songs and then delete them at the same time? Sometimes I create the song many times until it sounds good and then you have to delete each one individually

thin ermine
regal oyster
leaden marlin
#

I would like to like the voice, instrumental or the style not just good or bad. That is too vague…

violet heath
#

I can't "Confirm lyrics being replaced" in the replace section because it won't scroll down to the button to click. any suggestions?

Screenshot_2024-10-23_153107.jpg
frosty breach
#

Brainblast
Replace on parts would be neat if it were possible to create an additional fork if you were to go back to full. Perhaps a creat 'Part' button or something.

I am still swearing by 00:00 Extends

I would still love to see arrow key functionality on the playbar. Perhaps also in combination of a other input, perhaps holding shift or something else, so smaller steps. IE: 1s skips vs 0.2, (I noticed a trend with 0.2, rathern than 0.1 increments)

true heart
#

suggestion: a way for us to mashup two of our own songs via a new feature - or an extension on the cover feature? i know this has issues though with copyright and stuff, but it'd be cool to have that more at hand

clever harness
#

Hey I want That!

I want to be able to do a cool guy custom header on my playlist too plz. 🥺

Screenshot_2024-10-23_at_9.59.54_PM.png
sage mirage
spiral vessel
#

Covers should be free. The amount of credits burnt on just getting the exact same result back is frustrating.

modest frigate
#

-- -- -- -- -- -- -- --
We need a "liked" filter on the create page. We need many things in the UI and Suno does not implement anything.

eternal pendant
#

make it so trending can only be one per person and one person once a week so others can have a chance please

fluid eagle
#

2

clever harness
#

Hot Key for Snap-to-Start when working with Waveforms

I like the editor for infill, crop, etc. However, I feel it's missing (or it's not communicated if it exists) a simple hot key to snap to the start of highlighted window. I like that space is a hot key for play/pause, so it makes me want a double tap on space to snap to start of the loop.

tepid dome
#

i need to change some words on the song, maintaing the same readom, but on the selection REPLACE its create me a new track without maintaing the same readom

#

is any chance to fix tat?

raw mist
#

Suno team - you are amazing! Please get the vocals clearer.
Thank you.

deep onyx
#

All the upgrades have been super helpful, BUT

Can we please get instrumental covers for songs that have vocals rendered in?
The results are either the exact same souns WITH vocals still present OR a wildly different song output

lusty blaze
#

Can you please fix the Instrumental algorithm so that it creates songs longer than a couple of minutes? Maybe allow a length suggestion factor for those who creating instrumental tracks? I am trying to make music for longer format travel videos and creating 1-2 minute instrumental tracks is not cutting it. Extending does not work and is a super slow burn when it does as you have to back time your track so far and then the extension is literally just giving me 20-30 seconds of extending. Allowing us to make a 4 minute instrumental would not bite into your profits that bad.

rocky wigeon
#

Suggestion - go back to original video style from generation. It was okay the way it was. Why was this even changed?

tidal isle
#

Suggestion for a new function: upload an old song (created by suno) and suno sings/do it again. only in better quality. this would also attract an additional target group, namely amateur musicians or artists who upload their own old songs there and get them back in better quality.

tepid trench
tidal isle
final echo
#

it'd be cool if you could combine two similar generated tracks (like same tags, but only the instrumentals, not lyrics) into one, mishmashing their instrumental styles and rhythms

#

say you covered an audio recording and it generated two cool songs that both sound cool but might sound better if mixed together via AI lol

open pine
#

Bad idea: look up songs based on lyrics used

hearty star
#

Just here to put in my usual 2 cents ^^:

  1. Suno is becoming a really great service; keep it up! You've already implemented all my previous suggestions (so I guess I'll add more now LOL).
  2. Allow us to pin on the left any playlist that we want. This would remove the need to load the Playlist page (thus consuming 2 less clicks on our end and 1 ENTIRE page load per user per trip to find a playlist on your end).
  3. Allow us to Add the "Full Song" tag to cropped outputs 🙂 Sometimes we are cropping what becomes the entire song.
  4. It would be helpful to add a mousewheel zoom feature to the song cropping interface (the longer the song gets the harder it becomes to click the right location).
  5. Ensure that Cover songs don't lose their original Cover song attribute when a Cover is made OF a Cover song. For instance in my public Timbaland Contest playlist three of the songs lost their cover status because I covered my own cover.
  6. On Arch Firefox, the interface is pretty solid. However I do from time to time get the application exception error and black screen. On android Chrome the interface is laggy with up to 10 seconds before a click on a song responds. You can duplicate this yourself (as a dev) by opening my All playlist. I add all my good creations so it's up to 157 at the moment but it's been slow for a long time (probably before it was in the 100s). If you can clear this up that would be amazing. It seems to be an issue with loading the playlist (sometimes it all loads fast then errors out, sometimes it loads really slow in chunks and works for a while before erroring out). The android Chrome error is also an Application Exception (so it is not likely a single browser issue).
  7. Playlists. When they get long it becomes hard to scroll through to find a song or section of songs. What if we had the ability to add section dividers? for instance: ---Dubstep---, dub1, dub2, etc... ---Rock---, rock1, rock2, etc. ?
river pilot
#

horizontal videos, higher quality on the videos (pro and up)

opal spoke
#

Im following category,instead of show the profiles,it show some shuffled or filtered musics from who you following or follow you,it’s cool for listen more easily the other users songs.

jolly wharf
#

For CONTEST, like the Timbaland remix, consider creating its own showcase section. That way they're not mixed up with everything else.
👆

#

An "improve vocals" feature would be a great addition.

wise blade
#

A drop down window to select the pitch and another one for tempo before generation would be nice

#

Here is what I would suggest:

image.png
ornate osprey
#

The ability to include a Youtube link in the final version of an AI public song for Premium users that you can also hear (Embedded in Suno). This way it makes a better hearing of some AI songs with increased quality or final versions. Like if you agree 👍

compact hollow
#

Training a model on our own voice for vocals would be cool.

true heart
#

suggestion: opt-in option to see notifications for new releases from people you follow

naive solstice
#

Make functional when PRO/PREMIER users can judge public compositions from 1 to 10 ⭐ ⭐ ⭐ ⭐ ⭐ 1/2 = 1

  1. It's make know to SUNO what tracks is rly good
  2. It's give autors real feed back from other autors
copper valley
#

Longer uploads for sake of covers, as I’d love to hear songs I wrote years ago in different styles, although…. The cover feature does eat through credits where often you get nothing new every now and then. @compact hollow’s idea of vocal training would be interesting to see

clever harness
#

Extend / Cover Stems

Ain't no way a billion people haven't asked for this one, but if I create a nice melody with hallucinated vocals (by design of course) I can normally work around this with great success 😎 . However, when I can't, I want to be able to get the stems (I can do this in Suno), and then feed the stems back into Suno as a cover or extension (I can't do this directly for reasons I don't understand)

The workaround now is:

  1. Create a song
  2. Get the stems
  3. Crop the instrumental stem
  4. Work off the crop of the instrumental

Side note: doing this breaks the UI, but that's the price I'm willing to pay to get that sweet sweet extension menu option.

Screenshot_2024-10-25_at_6.41.05_AM.png
sleek abyss
#

We really need a creativity/adherence slider, at the very least for the replace section. It doesn't work as a way of fixing vocals if it just picks random lyrics from other sections instead of what we have typed into the change-lyrics section.

naive solstice
#

Credits slowly go down by self its Halloween feature or bug?

wise crater
#

Tell me how to solve the problem. This error has been observed for about a week on Android and Windows.

Screenshot_2024-10-19-00-06-34-098_com.android.chrome-edit.jpg
remote stag
#

My biggest hugest, most extravagant request.

Please include the Jersey Club genre for prompting 😭😭😭

I'm burning through credits trying my hardest to generate some.

Thank ye, kindly.

cosmic lodge
#

I wish Suno had a detect genre feature for imported sounds or actually any web link

wraith sequoia
#

excited Suggestion

Provide Suno creators with the ability to download all songs and stems (in WAV format) from their custom playlists as a single zipped file.

thorny pelican
#

When pieces are created I often cannot really judge, whether I like the outcome because it highly depends on my current mood but maybe for a different thing this could be useful.

Can we introduce a memo field (textarea) where we can store our current thoughts on that piece to make better informed decisions at a later time?

tepid trench
novel compass
#

Can we please get a checkbox for playlists that would disallow adding duplicate songs to it? Like a "[ ] Allow Duplicate Entries" or something like that?

real lynx
#

Id like to be able to set style weights. Some styles seem to dominate all other styles. For instance, emo rock is dominating everyother genre i put. I'd like style weighting like midjourney allows. Like, Emo rock::0.25, Trance::0.75

ornate osprey
inner jungle
#

I want to see key changes, when it comes to creating songs.

#

or a key changes prompt

nocturne verge
#

more control over the gender of vocals. seems like a simple thing to control. the wrong gender on a specific artistic idea can ruin an entire song. yes i have tried the negative prompt. doesn't work yet.

remote pond
#

I don't know why it does this, and it isn't exactly a bug, but, after I create two songs, it does not show me how long they are until I play the selection almost all the way through. This makes it difficult to determine if it will be appropriate for my work.

I could extend one that is too short, but usually, if I tell a piece that is 2:32 and I'd like it to be 3:00 or longer, I tell it to extend it starting at 2:00. And it only extends it 0:30, making it about as long as the original.

Even then, it will tell me that the song is 4:00 but when I listen to it, it ends after 2:30 and the rest is filler.

ornate osprey
#

let us choose the maximum minute/seconds length for generation and extend option, so songs can end better.

leaden marlin
#

Definitely adding a way to identify song key and tempo would be a great step in the right direction. If we can export sections to midi that would be the next level for sure. Lastly, the ultimate option would be to have a creator uplaod their voice as as a kit option to share or just a private model to be used in generated music.

covert ivy
#

I suggest...

  1. Let users make public playlists that are discoverable... (Kinda like Spotify Playlists).

Getting a few good curators for different genres would likely help a lot (may have to create a way for them to make money - maybe like what's done for Spotify eventually, with people charging $1-2 just to take a look. lol)

  1. The songs you make public should have human readable URLs... Would look better when sharing on social.

Maybe something like suno/username/song-title - which would complicate things if the user name gets changed later, but even so.

twin hinge
#

Suggestion 1:

  • An upscale function similar to Midjourney's Upscale Subtle/Upscale Creative.

Upscale Subtle reduces grain/static in the instrumental and vocals while staying close to the original track.
Upscale Creative works similar to the Cover feature but keeps more of the same lyrical structure as the original track.

Suggestion 2:

  • Built in mixing/mastering - a Suno built DAW w/ tutorials.
real lynx
#

Would like to be able to edit the song style in the description. When I use a lot of cover/extend/etc the style gets mixed up if I style change. I have some that sound nothing like the style tag it is. And two of my songs even say (No style)
I'd like to be able to set genres and define what the song actually is lol

fringe shale
#

Custom voices, please! I want to use one voice for all songs, not to get a random voice every time

woven fractal
raw mist
#

Please work on getting clear clean vocals. Currently they are distorted and non-professional.
Thank you.

frosty breach
remote stag
raw mist
#

Paid users should be able to talk to human support people. Discord is cool but so many posts go unanswered. Heck I'd even pay to get my questions answered. Thank you for considering this.

tepid trench
#

Where are you seeing this? I mean, do you see old prompts on the song page for a newly-generated song?
If it's not showing you the old prompts, it's not using them - as far as I am aware Suno does not remember what you prompted for previously and what's displayed to you afterward is the prompts that were actually used.

The only exception to this is if you're extending a song, in which case it'll use that song as context for the extension.

raw mist
#

Allow longer uploads than 60 seconds for paid users.

Thank you.

tepid trench
raw mist
tepid trench
raw mist
raw mist
latent ruin
#

Hmm. Got one. Grantable cover permissions operating on a whitelist. Like being able to press a button on someone's profile so they can cover your songs, or an option to turn the covers option on for individual songs in the normal menu for them.

weak sphinx
#

Add to scenes the ability to make Lyric Videos or by its self. Basically you can AI request a little cool background for the lyric video and your music's lyrics will appear

bold berry
#

jump to section or 1.5/2x for generations/covers

prime rock
#

Can I just ask what the reasoning is behind making the UI less and less functional? Now it’s not possible to see the up and down arrows in the create page because everything is locked and can’t be scrolled sideways, and the useless giant left third covers too much of the page to allow the song style to be visible. It’s also not possible to cover songs anymore unless you find the song in the library list. It can’t be done from the song page. Every time I log on it seems like something new is broken and the last thing that was broken isn’t fixed.

proper cave
#

IDEA: We should be able to specify the key of the song

patent nacelle
#

Should be an option to get refunded of our credits on a song by song basis. I probably burned a whole months of credits in covers because a lot of time the AI spits out the original, or does whatever it wants instead of actually covering the song.
or better : since it's in beta, with a lot of errors, it should be totally free for Pro and Premium users until the feature is ready to fully roll out

#

of course a refunded song would be deleted forever

woven fractal
glass portal
#

Suggestion: It would be nice to be able to sort the track order in a Playlist

frosty breach
#

I would still like to see this suggestion be revisited. Especially now with cover and replace. Having hard cuts to bits that are so different than the rest of the song would also be quite useful with these tools in mind as well.

I think the mistake in the first review of this suggestion was never from a users perspective. If my third part has 0 context it can do whatever next. But once we go to full the entirety is in context and will allow Suno to do some interesting fusions and mixes compared to the 00:01 extensions.

00:00 extension allow for even more drastic switches that 00:01 would never be able to, while it still can however. I want more instant changes myself and potentially then go with cover or replace to make these transitions smoother. Hell it would allow for easier rhythm changes.

So many uses for 0 context which is hard to put all into words without it looking like rambling.

covert ivy
#

I'd like an option to rate songs from 1-10 when I'm working on them.

The 👍 👎 aren't good for finding more promising variations that I might want to work on... Songs can be decent, enough to deserve a 👍, but not necessarily decent enough to continue with.

Not that I don't have workarounds. But it would still be nice if Suno enabled more complex workflows for when you're exploring a specific idea.

... I'd ideally also like to have the option to add tags about what I liked / didn't like about a version of a song.

lapis osprey
#

I couldn't find "general suggestions"...

I love Suno. It's very creative , super easy to use, and follows style prompts pretty well.

I'm an annual subscriber and would love to stay so, but the number one thing that will prevent me from renewing is the overall poor sound quality. If you can fix the clarity issue to at least as good as Udio, I'm in! Clarity needs to be #1 priority.

2nd to clarity is expressiveness... most tracks are pretty repetitive and can get boring... some greater expressiveness in instrumentation and vocals would greatly improve Suno's offering. Yes, there are some style tricks to improve this... but would be nice if baked right in.

Thanks! Can't wait for next version! Hoping clarity is #1 priority!

modest frigate
# lapis osprey I couldn't find "general suggestions"... I love Suno. It's very creative , sup...

try Suno V 3.0 for expression and expresiveness. Better clarity would be nice but it might be not easy to achieve. I would be happy at least is Suno would improve the UI and add some song managament. I use also Udio but there I get in 90% of gens drunk vocalist or beginner vocalists. To get normal singing and good melody in Udio is MUCH more rare then in Suno. I never heard drunk vocalists in Suno so far ! Are you happy about the singing performance in Udio? I realized that most Udio users are fanatics and they hallucinate about the singing quality of Udio. Sound quality is not everything in music. And I do know the settings very well and I tried them all.

lapis osprey
# modest frigate try Suno V 3.0 for expression and expresiveness. Better clarity would be nice bu...

Yes, Udio has its problems too. It does better with 30sec gens than the full 2min. It does allow tweaking for better adherence to lyrics which works fairly well. Creatively it does good with the right style prompts... otherwise it's a lot of wasted gens.

There's no excellent choice yet, but it's super troubling to get a Suno Gen I love but the quality is too bad to make use of it.

Maybe one day we'll have the ability to reimagine our tracks with creative upscales like you can do with images right now... that would clean up the instruments and vocals to sound great without getting too far askew from the original (or very far if that's what you request).

Suno is super fast... even faster lately (I've really noticed a difference recently). I'd gladly give up that speed if I could get just as good of songs out of it but with much higher clarity .

modest frigate
#

There is nothing like a "GOOD prompt" in Udio that has an influence on the vocals ! This is a dream/hallucination of amateur Udio users which don't understand the science behind the AI models. They try everything just to make it work better, but it does not work better if you add things to the prompt, it is an illusion.
I tested that intensivly.
As any other AI, Udio is in some genres very bad, bad of good or very good. This dictates the song quality the most and also the SINGING.
And this is the most important tag: the genre.
In some genres you get more normal singing and good melodies, in other barely something that you can use.
The rest of the tags choose only the mood, the pace, the instruments, the kind of voice. But NOT THE SINGING !
There is no tag and command for good singing, this is the problem of the Udio AI model. And that's actually the biggest problem of Udio: the singing!
Udio is not such a great singer as Suno ! You need to be lucky and spend a lot of credits to get a normal singing and even more luck and credits to get a good
vocal melody (you have also more credits than in Suno for this). But you get songs Suno cannot make so it is good to use both.

I generated 6000 clips in Udio and in only 200-300 the singing was normal and the melody was good in the same time.
Some of the clips were great though, music I wont never get with Suno. But I work with that afterwards in both AIs to develop /finish.

#

What you can do if you work with both AIs:

  • find out in which of your genres Udio or Suno is worse and work rather with the better one for composing (in R&B Udio is much better and in House music Suno is much better in my opinion). Finish the idea in Udio if you think this sounds better.
  • use Suno to generate a good voice melody and good voice and upload in Udio to continue there if you want a "better sound" (although the drums and hi hats are not so good in Udio despite clarity).
  • use Udios 32 sec clips that Suno cannot compose anyway and upload them in Suno, extend up to 2 mins in Suno (and than if you want upload in Udio again for finish. But the problem is that remix in Udio works only for 32 sec parts)
  • use both AIs to upload your own music and see which is makes a better version (and take less credits) and play with both by uploading what they generated in the other AI. Its is lot mor creativity using both.
prime jasper
#

I would be really nice to have the option to choose "Intrumental mode" in 'Replace Section' just as you have in 'Create' and 'Extend'.
Some sections have vocals, singing or choir, that I don't want.
It's really hard to remove these now. Extend changes the music to much. Replace section keeps it on trck much better.

scarlet vine
#

Under 'Style of Music'
break it int 3 sections.
Style - remains , but needs to allow Tempo notations (accelerando, Rallentando, Allegro, etc.)
Vocals - choose: male, female, duet, robotic, harsh, etc.
Instruments - choose predominant instrument: brass, horns, woodwinds, bagpipes, harp, etc.

Also a help UI on how to best use these sections.

  • would be nice if Tempo, Instrumental, and Vocal notations could be supported in the lyrics section, so you could specify for each verse if you wanted to change during mid-song. Just need to use standardized format so they can be recognized. (again a help section listing the choices available)
prime rock
topaz dove
#

what if you could enter a prompt and just get endless music?

#

id subscribe to that

#

seems to me thats where this is going anyway

amber finch
#

We need a "style of voice" seperate from "style of music"

mint iron
#

I looked but couldn't find this. I would love a way to link existing songs together. Kind of like the way we can extend existing songs, but this would allow for a short part 1 with a longer part 2, or vice versa (the piece I'm currently working on has a 24 second intro and in total is under 4 minutes long. With a 2 minute limit on extensions, I have to export the audio and stitch it together, rather than be able to showcase the work here on suno)

hollow locust
#

I can't be the first person to suggest this: Android app.

wet zephyr
#

I have purchased the Premier Plan, but the videos and music covers I generated have watermarks

amber finch
#

It would be nice to have a video gen that is just the lyrics so we can composite them inot our own videos. Would save a lot of work starting from scratch.

cold wigeon
#

I would suggest to add a separate prompt for voice and it gives you unique index that you can use in the voice prompt to keep using the same voice over and over

amber finch
#

the model really needs practice with non-standard stanzas. anything outside of 4-line stanzas ends up jumbling the lyrics.

slender socket
#

Hi I've been using Suno for a long time. Would there be any way to have a setting to choose male or female vocals and another setting for using the same voice from a previously created song? Sometimes I have to make 20 plus songs to find the same sounding voice. Thanks.

runic pier
#

Number 1 suggestion: Not everything generated should count as a song. Not everything should go to the "songs" tab. There should be more tabs in the library "songs" "instrumentals" "clips" (from replace section) this would be a good place to start.

tacit idol
#

@clear python Please don't use this channel to share your music.
That's what #🐣┃suno-showcase is foar. 🙂

violet sierra
#

If I could get a refund for every random piano cover...

cyan sable
#

Please make the "Following" list standard list, instead of huge gigantic grid of icons. A list would be smoother to scroll.

untold flax
#

It would be great to be able to make mashups of two songs and have a new song come out as a mix of both, combining their styles and where you can determine which parts of the lyrics of each one stay and how they are mixed. Also, there should be advanced sound post-production tools.

ivory rover
#

When you open the sub menu with the 3 dots on mobile and click something in that menu it clicks what's behind the menu too then usually opens that, like if you do it on the create menu it'll click one of the genre tags and open that instead of say cover song. Really annoying to use the mobile website when its almost impossible to get it to do what you actually want without trying multiple times.
I put this in here since it's mobile browser. Only other option is ios which I assume is that app so it doesn't apply.

tepid trench
vocal portal
#

Allow us subscribers to get rid of the Timberland advert that takes up the whole right hand side of the browser window. One of the perks of paying a sub is that we should not be seeing adverts.

violet sierra
#

Taking its time to generate album art. Can you make it so it does it in the background and allow an uploaded file to be worked on faster?

gilded arch
#

Ability make a song public on Suno without letting others extend/reuse?

modest yarrow
#

some 'dailies' and 'weeklies' would be nice just by providing tags that people are required to use with their own lyrics, would help with writers block having an instrumental already set up on their end or one provided by suno with the suggested tags

it could be as simple as making a discord channel that people can post their songs with only those tags but with their lyrics and others can listen to see what flows well (which would personally help me learn more about how a.i responds to our lyrical inputs)

edit: not sure if this has been suggested before, dont want to steal anyones idea but learning tagging through some means would be highly appreciated

proud slate
#

Please optimise the browser experience as long playlists crash the site with an error and the more you scroll the slower the response time is (perhaps not loading everything except what is in view is better)

Also there is no way to play a playlist on repeat and no way to change a playlist’s picture.

Songs in a playlist should continue one after another, but just because I click in a weird way, the next song played is a random one (maybe another song from the same creator) and not the next song in the playlist

sick lion
#

A radio station that plays songs from people you follow, but with options: most popular, new and others (instrumental, genres). But mainly a radio that shuffles following songs.

wicked nymph
#

I think it would be cool if we could turn notifications on and off for profiles we follow, so we could get alerts when they release a new song publicly. Just like how YT notification system works (where you can choose whether or not to receive them).
I also think it would be nice to link a few links directly on our account, since many people do this by putting links in lyrics or on the profile. It would be great to have direct support for this on the profile, like in the screenshot below:

image.png
wary coral
#

When using "Get Whole Song" on an extension, there's seemingly zero indication the process is running, I had to manually refresh the web page to see the newly generated song being generated - I was convinced it hadn't worked, and was about to press it again. for a plethora of reasons, not ideal, including but not limited to customers spending credits they didn't intend to.

young bloom
#

Genre voting, like what distrokid has, would be great for creator visibility

wary coral
#

Link on covers back to the original, fine if it's in the three dot context menu, just want to jump back to it quickly.

flat fractal
flat fractal
#

A way to favorite and/or feature our favorite song creations. A way to promote or suggest our "greatest hits" when visiting our profiles. Well, I guess most plays from different viewers would make it a greatest hit kind of thing. I'm thinking more like an editor's pick. Artist's choice. While any chosen piece by any given artist may not resonate with others as much as it does personally for the artist(s) involved in creating the piece it does speak volumes about the artistic personality; what the artist feels represents their personal style that they like to bring so that the listener may also know what to expect from them.

modern summit
#

would be nice to have a note section or something so we can eventualy write a litle description or just our thoughts about a song, why we created it, what is the context, etc. sometimes there are songs i'd love to share but i'm hesitant because they are also pretty personal too or i did them for a reason which i'd like to explain

wary coral
#

Add a badge / indicator that a song was created while subscribed and allows commercial use - don't make a user manually math out / remember when they were subbed, simply add the badge to any able song.

sonic tiger
#

This system has the potential to generate 'flawless' songs, but it still has a few issues that need attention:

The support for 'dangdut' doesn't always work accurately. Many songs end up being rendered with pop or slow music instead.
Several songs are too short, falling below the 2-minute mark. These truncated songs require extension and final touches. While finalization doesn't take up any additional credits, the extension process consumes 10 points, meaning that two songs can exhaust all daily credits. Moreover,
It would be preferable for the song generation to start with a natural length of 4-5 minutes instead of just 2 minutes.
The adherence to [meta tags], which describe chorus, verse, etc., is inconsistent.
Sometimes, even after extension and finalization, the length of the songs remains shorter than intended when summed.
Additionally, there have been instances where the audio fidelity is compromised towards the end of the song.
#

iwould lise rto think of the Suno 4 when t willbe available

mint wharf
#

Downloading a MIDI file of the songs made can be awesome so we can rework more precisely the song and/or change the instrument, example : recently I made piano solos but the quality of the piano can be bad for multiple songs, if we can have the midi file we can play it with a better piano plug in to have a better piano quality. I tried with FL studio and other software to translate the audio file into MIDI but every time that was a completely mess and it’s impossible to work with it.
Being able to change the instrument can be good though, if I made a piano solo but I want another piano for a better quality or different sound etc… it can be another great option.

vagrant wadi
#

With the mobile site for android, I have a few suggestions:

  • allow playlist images to be replaced/added/commented - for me playlists are like albums.
  • let people comment on songs or share other people's songs
  • add an embed share feature so you can add it to your site or to social media so it looks good rather than a big long link
  • allow the app to stay in background longer so if you're listening and playing a playlist but the app is not in the foreground or the screen goes off it doesn't interrupt after a while.
  • fix the issues I've seen elsewhere with only a single song playing and then it stops or can't find the next one. This happens for me in library or in "create" mostly.
  • fix the "clerk.suno.com" redirected you too many times error - which happens to me on chrome and firefox every time a suno page loads.
tulip fox
#

Make the like button available after 50% of the song has been played. That will stop these "like groups" flooding songs with likes without even listening to the song.

scarlet vine
#

Suggestion: If I leave the musical style section blank, please have Suno A.I. fill in what styles it ends up using rather than leave it as (no style)

  • this makes it properly classed into the proper style for search, and also provides a prompt if I want to duplicate the style in the future.
quasi talon
#

Suggestion : folders to directly send all the musics generated in it. This would allow us to restart generation of an older music without looking for it in page 86.

mental helm
#
  1. As some others have mentioned, it'd be nice to have the ability to re-use the same voices, for consistency when making an album, for instance. 2) Other than adding to a playlist, it'd be nice to add to an Album, that would eventually get uploaded to a music distributor. 3) Would Suno AI ever want to offer a music distribution service? They could make extra money that way. 4) I want to be able to force Suno AI to accept words and sentences I want to try after the first initial song is made. I have to go through a few different rounds sometimes until Suno uses my words, wasting credits. Sometimes it never uses my words. Frustrating. ---- But overall, I'm loving Suno AI, thanks for existing!!!
gaunt tapir
#

J'aurais une suggestions a proposé pour le site web & les playliste.
Pouvoir directement généré dans le dossier playliste les chansons comme ca c'est directement trié.
Et si on veux travaillé sur plusieurs album a la fois c'est plus facile de ce retrouvé haha

hidden spindle
#

Dear SUNO Team,

You’ve created an impressive sound development platform – one of the best AI platforms out there, right alongside ChatGPT and Runway. However, your marketing actions often feel off-putting. The competitions are so brief that only a few can participate, and in the cover contest, we had to work with a rather uninspiring song. Now, Timbaland and Alexandro are taking up a lot of space on the platform, which seems unnecessary. Sometimes, it almost feels like your marketing team is achieving the opposite of what you intend.

One more suggestion: I’d love to have my songs automatically sorted by newest first, instead of always seeing the most liked ones. The filter is available, but unfortunately, it can’t be set as the default.

SUNO is truly a fantastic platform – keep up the great work, and please reconsider your marketing actions in the future.

Thank you and best regards.

proud slate
wary hazel
#

What are the best settings in Suno AI to create the sound of soldiers singing during a march, where one soldier leads with a line and others repeat after him, similar to military cadence chants? I’d like this to be in an acapella style, with no instrumental backing. Is it possible to achieve this effect in Suno AI?

wraith zenith
#

I would love to see some post-processing tools available in Suno, like equalizer, normalizer etc. It would make the entire experience whole without needing extra tools

prime rock
#

It would be nice if you could select a song to cover from the create page

prime rock
winter hound
# amber finch the model really needs practice with non-standard stanzas. anything outside of 4...

I have little experience with vocals, except mixing IRL; I compose instrumental music. But FWIW, I currently do a song with the 6-line verses. Sometimes Suno splits them as 4+2, a little more often 3+3 (with a little of the pause in between). Most often, though, she makes 6 into a solid verse. If this helps, I have assonant rhimes and the last line is one syllable shorter than the rest. I agree, I want a better : sometimes she breaks the lines as she fancies, and, although the generation goes into the trash, indeed, I might disagree a little on the characterising this as a jumble.

winter hound
#

Please please please do this little change to the song trimming resolution. Currently, it is 200ms, which is like the eternity. If one cuts too early at, e.g., 1:23.04, it often becomes too late at 1:23.06, while there may be a perfect place to cut right between them. IRL, I cut in WaveLab or Audition with a sample precision, but usually 5-10ms is shorter than enough. I propose leaving the number set by dragging a marquee as it is, but not round the number to the same 200ms when it is typed. Maybe 20ms for find a more precise place by the bisecting of the interval, or even better 10ms. I reason, unless the AI generation is involved in the trimming (apparently, isn't), or unless the model requires the length which is invariantably a multiple of the 200ms granularity in case the trimmed material is later extended, this seems to me a very easy change. A ~15-20ms fade at the typed cut point, and one would be able to make wonderfully precise, studio-grade cuts. Currently, I can only do, to Extend with v3 and the lone tag, [end], which takes a 6-8 generations, and twiddle ±1s with the continuation timestamp, a half of these usually refunded. These, the refunded, do the cut, but often with a buzzy tail in the background. A cut is the sometimes preferable one, when the UV is low or the total silence.

And maybe the other day, a precision cut throwing an intervall out from the middle of a song will be! That would be a monster feature! Especially given the propensity of Suno to repeat the whole verses or the parts of the verses… But it is useful only with a finer cut precision, at most 10ms, unsure of 20ms.

amber finch
#

It would be nice if credits didn't expire. I hate having to mindless grind out songs or rush to use them all.

hidden bison
#

Yes , please make it so you can control the voice better, male, female, group of, style, tone, pitch 🙂

#

and slightly less autotuned

feral aspen
untold flax
feral aspen
untold flax
modest kettle
#

Why can't I search history guys?

warped ridge
#

what's wrong with the "cover "?

I find its output is much worse than several days before.

It can't detect the rhythm I sing or upload!

amber finch
#

folders for playlists for organization. I keep playlists of attempts and for better gens seperate and it would be nice to be able to group these to gether and organize them

jovial bridge
#

I wish suno had messages to interact with others

latent ruin
#

Notification whenever one of your songs or playlists hits Trending.

compact hollow
#

Rollover credits for at least 2 months

jaunty drum
#

Can I recommend getting rid of the prompts displayed in the finished songs, as some are very misleading. Some songs aren’t even close to the same genre as what the prompts suggest

#

Also I blew through like 140 credits getting rock songs when prompting for rap 😂

unique ledge
#

for the custom audio upload option: can you make a option where you put original lyrics and your own lyrics separately and the ai will make the exact same song with the same melodies and choruses and voices as it was except with different lyrics

floral sandal
#

"regenerate video" still doesn't actually do anything 9 out of 10 times

raw kayak
#

Put at least one social network link

#

It would be nice to be credited again via a button when the creation is off-topic in relation to the request made, or when the creation is less than 20 seconds long, as this is unusable. With my Plan Pro, I lose as many credits for off-topic and too-short creations as I do for completed tracks.

magic granite
#

More control on the nature of produced audio. For example Covers? It would be fantastic to have a slider for how close the Cover should follow the referenced song's structure, maybe another to keep it close to the original sound/instrumentation. Etc, etc.

desert birch
#

coming in future Paypal payments? I mean that is better than creditcard 😦

brave ridge
#

a simple one.. a folder structiure so I can list songs by project

indigo ginkgo
#

I'm going to make this request again, because I truly think this is an easy and important change for a lot of the creators on this platform:

It would be really nice to take some of the stress from the trending system:

  • Please increase the amount of trending spots from 35 to something like 60: More people being able to trend is always a good thing!
  • Please limit the amount of songs a user can have in trending to 1 or if you want to be nice to some people 2: This opens trending up to more people. (Just bump the other song, if a new song enters global trending)
  • Please make songs that trend in global, disappear from English trending: This allows more people to reach English trending at least and opens up more spots for people in total.

I truly think this will alleviate a system that has now gone crazy. When I started with suno, 10 likes would get you into the trending page which had 50 spots. Now 100 likes won't even get you to global trending. And even though trending isn't anything more than a good spot on the website, it matters a lot to all of us mid to large tier suno-artists and to be honest when I started and got to trending it gave me a huge kick and fueled me to create many more songs.

As far as I can see, none of the things I am proposing have negative effects, maybe user limits, but more different people in trending is truly a good thing and I say this as someone who is still able to get there regularly. Even if you only increase the amount of spots it would help for now. Is it the perfect solution, definitely not, but it would help us out currently! (so change that number 35 to 60)

loud kite
#

Personas is cool but for a persona 2.0 being able to mash up the essence of two different genres or songs would lead to interesting combinations

serene charm
#

It would be amazing if one could make the AI sing with a clone of their own voice!

primal loom
#

You should focus on improving the audio quality before creating features that, in my opinion, are useless. Sorry, just an opinion...

robust jewel
#

Make Persona, like Cover Song or Get Stems should be marked wish the ? that informs users they need to be subscribed to use it. Currently, it just doesn't work without an explanation.

sweet flicker
#

add a discord channel for personas 🙂

echo surge
#

Personas: I'd love to see:

  1. The ability to apply a persona to a COVER or EXTENDED song. This is where it can shine, as it can take a style and cover a song you already love
  2. Personas to be private by default (or at least some option to have this as the default preference)
    Thanks.
runic pine
# serene charm It would be amazing if one could make the AI sing with a clone of their own voic...

You can kinda do this if you upload a good recording of yourself singing. I've done it a couple times and it's pretty difficult to tell the difference between actual me and the AI voice. Example: https://youtu.be/QAg-0A3QfeI

YouTube
The Redeemer Project
Pie Jesu

This is a sneak-peak at one of the tracks from me next album, titled "Requiem Ex Mortis." The album is a full-length Requiem mass, entirely in Latin (except the Kyrie, which is Greek), and in black metal (mostly).

This particular track is NOT black metal. However, it is special:

I created this using my own vocals. That's right - the first "pi...

▶ Play video
sharp eagle
lapis summit
#

The line spacing with adding the Persona adds a little more than I'd personally like to each song, especially when you have max styles / excludes.
Maybe add it at the end of the title line instead

twin hinge
#

What good are all these new features when the UI/UX is still so buggy? I'm sure as a community, we can agree that we would much prefer you to fix all these bugs first. THEN add more features.

royal venture
#

Let us download SRT files for the lyrics, time stamped. You must be able to do this, or else how you matching lyrics in video to song.

marsh steppe
#

Suggestion, let us be able to cover the same song but with reworked vocals, cause sometimes I have an amazing beat, but the vocals are just awful sounding

next ridge
#

I don't know I can mostly deal with the UI new features are very welcome. But when you do get time would love to have folders or a star rating, so I can sort them easier. (Takes so long to add them to playlist)

#

Also sorts (newest, most liked, most played, etc..)

amber finch
#

I like the persona feature but it woudl be nice to be able to build one from direct uploads. was hoping to use my own voice for the songs. covering the uploads to try to get my voice into one of the songs isnt working

misty blade
#

Hey Suno! "Covers" and "Persona" are amazing additions for creative control over the songs—I absolutely love them. Thank you so much! It would be fantastic if these two options could also be available in "extend" mode, acting as additional control parameters that could be combined into existing songs while extending, not just for new ones. Thanks again!

topaz mountain
#

It would be great if you gave all artists whose data you trained off of an attribution share from every output. That’s what we will be doing and all other ethical ai music platforms. I really like your product it’s just too bad that you aren’t ethical. Maybe that will change!

amber finch
#

It would be nice to have the ability to automatically add generations to a playlist in the prompt, so they dont have to be added manually. especially useful if you are gening a lot of tries in succession

normal wing
#

I have around 60 + pages in my library yet, no quick way in the menu icons, to get from recent tracks to the oldest ones . Also, I use playlists alot and can have a song in numerous different playlists, it would be nice when you click on a track, you can see what playlists they are in. It's a pain to keep going back into playlists and manually checking if it was included. Cheers!

rotund elbow
#

Can I subsequently separate the instrumental from the vocal in Suno?

blissful sail
#

First of all, I love personas! But they need to be restricted just like extensions and audio inputs should be (see my song Eye to Eye on that topic). With personas you can steal someones song by extending it, taking the persona and remaking the song with that persona and lyrics. I tried on a friend’s song. Until this is fixed, I have removed almost all my lyrics to prevent this. Please fix it Suno 🙏

blissful sail
inner shore
#

add comments

acoustic pike
#

I have noticed that the female voice sounds good while the male voice becomes robotic. Can't you fix this?

stone steppe
#

Hi there. Made the suggestion in chat earlier.

Would love an idea of having our own styles with the existing models at hand. Feeding it our own data/music to influence its output and possibly giving them names similar to personas.

I can definetly see potential concerns with this, it may have already been considered since I cannot imagine myself the only one coming up with this. I would love to hear about any updates on this if its being considered because it could open even more doors for Suno

Thank you for readingvulcan

serene charm
wooden vector
sturdy pike
# blissful sail First of all, I love personas! But they need to be restricted just like extensio...

This is the exact reason I don't put things public anymore since covers came out.
It's nice they restrict these things to your own library, but for the persona's the public/private setting is a joke while the 'extend backdoor' still exists.
It seems like a simple thing to implement considering they already have similar resitrictions and toggles in place, and I really don't understand why this is such a difficult thing to do. Especially with the legal implications and copyright.
They are selling us exclusive commercial rights if we are subscribed but they keep a feature that promotes and enables people to steal your songs into their own library. And now, personas.

hardy ledge
proper solstice
sturdy pike
strong shore
#

Hi there,

I’d love to see more options for personalization during the process of generating something specific. For example, after multiple generations, I could "like" certain tracks and then refine them to align more closely with what I have in mind. It would be really helpful if the model could take my favorite ("liked") tracks into account when generating the next iterations, using them as a reference. I saw today’s update about Personas (great update, by the way!), but my suggestion goes beyond just style or voice—it includes melody, beats, and other elements too.

Often, when I try to create a specific melody, like after uploading a recording of me singing an idea, I get close to what I want. But when I try to develop these tracks further, the generation often diverges from my vision. Since there aren’t many parameters to adjust aside from the prompt, it’s challenging to steer the process in the desired direction. I think it would be great if the generation could consider my liked tracks while using functions like "Cover" or "Extend." This could help capture the emotions, styles, or vibes I describe in the prompt, without altering the melody or core structure I want to build on.

I hope to see something like this in Suno AI soon. 😃

thorny compass
#

A cool feature would be that we would be able to sell our music

vocal belfry
#

Okay hear me out...Cover + Persona at the same time.

thorny compass
#

What also would be some good ideas in my opinion at least, would be
One: the remix feature but what I mean with that is actually you change the style of the instrumental, but the vocals keep the same.
Two: being able to upload samples of your voice for the persona

fathom tinsel
#

Add comment section on the songs and show your following on your profile as well

silk fable
#

Dear developers! It would be great to be able to loop entire playlists to play from the beginning after they reach the end.

wanton rover
#

The persona feature released yesterday doesn’t add much value; it still doesn’t allow you to choose your own AI voice when creating an album. The voice changes after generation Essentially, the persona feature just acts as a bio for the voice used in a song you've created, without offering any real customization. Suno needs to improve—we want the ability to use the same voice across different beats without changing the vocal style. In reality nothing has changed.

rugged herald
#

I’d like to be able to share Personas with another Suno user without making it public. I work with someone who has a different account and we have all these characters and both write songs for them. I would like to be able to work jointly without worrying about someone else using our personas.

thin ermine
rugged herald
clever briar
leaden marlin
#

SUGGESTION: Allow comments on songs and DM's to the creators if creators choose to allow them. I think it will grow the community without having to jump to Discord (no offense) to communicate. Also, could allow for creator to check mark a song to be considered for the SHOWCASE rather than spamming Discord (no offense).

obsidian peak
#

I love the persona feature, would be better if I could upload music ive already completed through other means to use that music as my persona. Ive got 50 songs I would like to upload and have the AI use as a template.

grand pulsar
#

After zillions of generations, I would love to be able to organize my songs into folders/subfolders and to rate them (5 or 10 star system)

barren violet
#

persona creation should not be defaulted to public

#

so far we know persona works .. and works well , as long as you only have one persona created ... more than one will lead to it auto-picking which one it wants to use .. that needs to be fixed .. and the search , only allows around 15 to be seen ... what if i have 30 , and the one i need is at number 20 .. how do i get it ? you cant .. we need that number fixed to allow infinite

hearty sand
#

why was the visual layout of the video download reduced in size? now it is practically impossible to read the lyrics on screen as they are too small. I appreciate that the stye tags were removed from the image, since they were always cut off anyway, but reducing the size of the cover image, and reducing the size of the lyric text made the video file less useful.

rugged herald
#

OK so I tried sharing a Persona link while on private and we can access the link but it won’t allow to use the Persona because it’s private. It would be really nice to have an option to collaborate with someone else without having to put Personas public. I don’t want the whole world to use my stuff but I still want to team up with someone else of my choosing.

royal venture
#

Once again saying to block the ability to extend, cover, or use any ones song without requesting access to the song from a creator 1st, like literally a user has to click a button requesting to cover or extend your song, and you get a notification with accept or decline button.

This new persona feature is once again another new way to steal creator content. Users can now make personas for all your songs and recreate your music, which they literally would be unable to do before.

Once again asking to obfuscate urls for embed, song, images, twitter embed link, download links, etc which are visible in the html code on the songs web page. Everyone else can do it, why cant you. Have a temporary download link generated when requested, not a static url.

This is not only the best option for your users, but the best option for you to prevent people from literally stealing your entire website, its code, and all the publicly available songs on your platform. Which btw has already happened in China and other countries. Yes I've seen them, and their websites are better. Not doing a very good job of protecting your content, which you would own when we all die, by letting others steal it all and put on a platform you have no control over.

cursive stump
#

Me in 08/20/2024 requesting a "persona" update

cursive stump
barren violet
#

i dont get why when trying to create a persona , some songs are acceptable , but other uploads arent ... since they all came from the same vocal creation of my voice

#

can someone explain that to me ?

twin grove
#

I see the natural expansion of using Personas to do Covers is pretty well-represented here, but I'm adding my own vote for that to become a thing

wild pumice
#

It would be good if license info for a song can be set. That could be used to differentiate licenses for songs created by free and payed accounts but also for payed accounts user should be able to set license and copyright info for songs they make public, so SUNO and other users would know if and how those songs can be used.

royal venture
# cursive stump But the songs aren't available to the public unless you mark them as "Public"

Im not going to argue this bs excuse. Public doesn't mean your free to steal. This is a platform built on sharing music. Making excuses for these devs being lazy is ridiculous. Whats the point of making music here for no one to hear.

Also just cause your song is private doesn't mean it cant be stolen. I can find your song through google on your profile if you ever post it anywhere. So you going to make music for no one to hear ever. Give me a break.

wild pumice
#

Setting up a license is common for created works. It does not mater how song is distributed, what matters is license. One may decide to publish as Public Domain, some as CreativeCommons or have all copyrights reserved. I do not see why would anyone object possibility to clearly designate license for a published song.

The first issue I see is that everyone stars using free account. SUNO preserves copyright to songs created under fee account. Then when user switches to payed account he has the rights. But in library, one cannot distinguish which sons are crated with free and which with payed account.

Other issue is when user makes song public other can do whatewer they want with shis song because SUMO cannot know license for a song and does not empower restrictions. If license is set to allow derivatives, SUMO coudl allow others to use those songs to creaste covers or personas or wjatewer. But if license does not allow derivatives, then SUMO can prevent other users to make copies.

solid sedge
#

hell no copyright rules

twin grove
#

Simpler idea: Playlists for Followers ('Fans') Only

radiant lynx
#

Please add the voice gender option all my songs are coming out female voices

wraith wasp
#

Seconding wanting CC support, I would like to extend material covered under CC-SA licenses in my own work and right now I would be violating that agreement doing so. I also just like giving sample creators credit.

marsh lichen
#

It would be great (maybe in some distant future) for users to do additional training of Suno. For example, it does not know what a balalaika is, so I'd like to upload files with balalaika sounds with the description what it is.

cursive stump
#

Uh, just download the song and then post the mp3

wraith wasp
#

Sure, but if they also want have a social element to the site it makes perfect sense to open it up to musicians that work with and promote creative commons.

sterile kernel
#

the ability to use persona for a beat, but name it something different. like same bpm or vibe or energy

spiral vessel
#

Sometimes extensions are good but too long. Just like there is a crop function now, could there be a fade out edit function, where you can fade out the song over a certain indicated range of the song?

oak sapphire
#

Shout out to the Dev Team for putting in PERSONAS.

Regarding: PERSONAS... I put this idea in Discord suggestions chat on Sept 20, 2024, (and I can't be sure if I was the only one to suggest it), but the Suno Dev Team did it. AND... I thank you very much!

bright dirge
#

So, I wanted to leave some feedback on persona, because its.. almost.. amazing.. My only complaint, is I can't seem to make a song over 3:20 seconds with it, it just cuts off at the point, some times shorter.
But, What I like about it, is Suno always kind of sounds to "perfect" but when using persona, it seems to dirty up the sound a bit, specially on rock music, which, for me, makes it sound better.

Another thing about person, Which was the coolest thing when it happened to me, I was working on a companion song for another song I had done so I was using persona, The original song was a heavy rock/metal song, but the companion song was slow, melodic, piano and strings, and it was perfect, but the song I was doing wasn't a typical song, it was more a less a long rhyming story, no clear chorus, verse, etc. and that was on purpose. to break up the song and kind of give it a gap in the middle between lyrics. so I just dropped a "Break:" in it. and it did the coolest thing.

Slow, piano and string driven song, broke out into the heavy riffs from the companion song, then right when lyrics picked back up, went back into the slow piano and strings. then at the end of the song, instead of fading out, you get hit with a quick drum slam then it started the intro of the companion song. EPIC

If there was a way to do that on purpose, that would be amazing. in 150 generations, that only happened twice.

void spire
# radiant lynx Please add the voice gender option all my songs are coming out female voices

This would be awesome! What I‘d like to see would be the option to Switch Personas in a verse like this:

[verse 1, Persona1-male]
Bla bla bla
Bla bla bla
Bla bla bla
Bla bla bla

[chorus, Persona1-male, Persona2-female]
Bla bla bla
Bla bla bla
Bla bla bla
Bla bla bla

[verse 2, Persona1-female]
Bla bla bla
Bla bla bla
Bla bla bla
Bla bla bla

Would love to see something like this.

Another feature I‘d highly appreciate would be to extend to another song, that Suno just mashes up the beat and the pace so that songs could flawlessly transition when you make tracks that are connected together (like an intro and the first track of an album or a transition between skit and track)

My 3rd wish would be to intelligently crop songs, where the AI decides what would be a short, suitable end if you slice a track. More than often the last few seconds (sometimes minute) is absolute gibberish and the beat continues to play, so you have to recreate another section and probably slice it afterwards.

strong shore
#

it would be good if Suno AI could do some editing and Mastering functions, like cleaning, pitch changing, reverb etc....

tulip fox
#

Select multiple songs and add them to a single playlist all at once.

rain lantern
#

Song mashups where we can match up our songs by using two or three different songs to create a new and unique songFuuuuuck

wraith sequoia
#

🗒️ Suggestion

It would be great if Suno has its own distribution service that would allow AI Music creators to push music to stores like Apple Music, iHeartRadio, Spotify.

rain lantern
wraith sequoia
# rain lantern This is not difficult to do on our own

Understandable.
For just a bit more insight on the suggestion -

If Suno or another AI Music Generated company were to have its own distribution service- they would gain several strategic advantages and could create new value for AI artists using their platform. Uploading AI music is already a challenge on a variety of platforms, and some AI music creators have experienced disappointing setbacks. If Suno or another AI Music Generated company were to do so - it would give them better control over how AI-generated and artist-created music reaches audiences, ultimately creating a competitive advantage in the music industry. Creating a new "kind" of artist. The first wave of AI Music Distribution.

Just an idea. thumbs_up(or suggestion rather)

rain lantern
#

I write & publish my own books on the side and the publisher I use does a very similar thing they will submit your book to multiple places such as Barnes and Nobles and Amazon

wraith sequoia
rain lantern
#

Maybe we should write lyrics for a song about how SUNO should you just this

wraith sequoia
rain lantern
#

It'd be nice if they implemented my suggestion as well

Also I would like to see their mobile app supported on Android and not just iOS

spiral vessel
#

Can we stop Suno from generating lyrics about 'neon', 'coffee' and 'city' 😆

zenith rivet
#

Add personas in the search tab!

grafik.png
mystic tangle
#

shuffling playlists 🙏 (you'd expect this feature to be in any music app, I was surprised by it's absence in Suno)

crystal atlas
#

Suggestion to memorize the new/top tab option or option in settings. I'm just tired of switching to the new ones. And I watch them most of the time.

thorny pelican
#

Please allow us

  • Filter for disliked ones in Library
  • Move all disliked ones into Trash
  • Empty the Trash with one click
#

And move the left area into a top top-most header and hide all the menu options in a single menu (hamburger button)

#

This would give us more space in the creation panel

modern summit
#

a little bunch of suggestions here:

  • make more things keyboard friendly, such as the public switch or the replacing feature which forces us to use the mouse even if the timestamps are visible
  • add the download option when you select a bunch of songs in the library
  • put the disliked generations directly in the trash
  • maybe a kind or reason/reward for us to clear the trash, like one credit per deleted songs or so, cause otherwise it's just there and just gets filled more and more and more. is there even a limite for it?
modest yarrow
eternal pendant
#

can you get this claim invite crap off my screen im already in the usa with the app

viscid cloak
#

The website right now is so cluttered...
Can we go to a design, like in march?
In March we had a full sepperate site for new and Trending

Split the Sites in more Under-Sites that you can select at the top:

Main Site
New Feature Showcase
Follows
For you
Artists you may like

Contests
(Right now) Timbaland
Past Contests
Flosstradamuss etc.

Trending
Global Trending
All other Trendings
Hall of Fame (Top Songs of Suno overall)
Suno Showcase

Browsing
New Songs
Top Categories
Genres

To give New Songs more of a chance, it could be its own Category with endless scrolling, would be cool!

It would reduce the clutter a lot and people can pick what they want to see
Also possible to make some Stuff on bigger Display

sterile slate
#

"Sessions" Each time I make a song if all the songs genned could be put in a category. It would make it easier to go through the billion gens I have.

If collabing I'd love to be able to put my profile or other profiles on the songs. Allowing downloads of collabs for both people that own the song. (Maybe it'd look like personas?)

Ik we have playlists but those seem to be best for albums!

open pine
honest bridge
#

Suggestion: "Notes from the editor" display section where users can add notes about their intent, inspirations, and neat tricks they might have gotten lucky with for each song. It would create more personal connection with the music from the editor to the listener imo. It would also be a valuable source of information for improving the models.

dusty urchin
#

whats the point of uploading my audio then making a cover of it when it cant fallow my tone and pitch. the next update should be suno reading a music sheet

cosmic terrace
unkempt zealot
#

Ability to re-edit/change/remove genre music tags

sick hollow
shut dagger
#

once again i am here to ask for project folders - so you can create a new folder, make all your tracks and have all the edits in the same place, then move the final song out and trash the folder~

hollow locust
#

Means to make it more musical theatre friendly... Currently really difficult to have a back-and-forth between two dif. singers. This person sings... that person sings... Chorus line joins in... You know, something like Skidrow from Little Shop of Horrors?

void spire
#

Personas on extend would be awesome

alpine spear
rugged herald
#

Using two Personas in a song to make a duet!! Then we could use tags to say the Persona’s name of who’s supposed to sing which line!

amber violet
#

This suggestion is for both Browser and iOS -- but in the playlist, highlight the whole row of the song green or something to make it visually stand out that it is playing. In a big playlist, I sometimes have no idea what song I'm on (trying to see the white lines bouncing is very difficult -- especially when it is on a black and white image!!!). Thanks!

honest bridge
#

Suggestion: When adding a song to a playlist, display an indicator (e.g. a colored a dot) for the playlists the song is already in

wild pumice
honest bridge
#

[TLDR: GraphQL for organizing music by queries] Organizing all of this material is a huge deal. With Suno we can pump out high quality song iterations all worthy of further iteration, but it’s way too easy to leave those in the dust. This is my journey towards a solution to clear that dust:

My solution began with a single WIP playlist that quickly grew unwieldy, so I split that out into music genre WIP playlists, such as “Textured Electro WIP”, but even those can quickly become dominated by a single song. It’s become apparent that each “bloodline” of a song often times deserves its own WIP playlist.

So I plan on building my own GraphQL powered "Suno Organization Studio" (for personal use, no plans to release it - other than to send it directly to Suno to consider as a proof of concept for future development) so I can tag all of these songs any way I want and then make powerful queries to get the song ID I need to just search for it (it would be great if this was just built right in). Some of the GraphQL queries I have in mind are:

  • Existing playlists for current song this would basically become obsolete with GraphQL except for public radio type playlists)
  • Song “bloodline” tree view, including all ancestor iterations (which can be opened to display further tree views) as well as all descendent iterations (in cases where you reuse a favored iteration to create many more iterations)
  • Ability to name “bloodline branches” - i.e. “song intensity increases” or “fade to ambient rest”
  • Queries like "return <all songs> from <some bloodline branch> that <have been cropped> and <are shorter than 1 minute>". Queries like this are cheap and safe (saved queries, enforced data types) to do with GraphQL, and work great with UI elements like facets
slate roost
#

Persona - The issue with this is that, it cuts songs and also restricted to a certain length based on the song the persona derives from. At times, it matches the length but if the song is slow, it won't complete it. It works well on male voices, will come back to give my feedback for female.

unkempt berry
#

Some at fb complain about the trending and like system. There is a video from Patreon CEO on YT explaining why this system does not work anymore.

umbral grove
#

someone needs to increase credits from 50 to 100 or 200

prisma crescent
#

When extending a song we have the buttons "clear" "Extend" add one more called "extend and end Song" PLEASEeeeeeeeeeeeeeeee 🙂

dapper anvil
#

Would be great if there was a way to use the new persona feature in tandem with the cover feature. Currently it's an either or situation and pulling the lyrics from a previous made song to use with a persona isn't even close to guaranteed to sounding like the same song.

stone steppe
#

Hello Sunovulcan

I would want to suggest a function that might provide Suno's already great generations even more degree of customizing. My suggestion is to enable a "Style Selection feature" where users could choose from their existing Suno songs' styles to generate new tracks in the same style. This would allow creators to produce music that consistently aligns with their preferred genre, vibe, or brand while also exploring new creative directions within a familiar framework. the "get whole song" feature maintains track consistency well. however, by learning from a user's larger body of work, the ai may be able to create music for each vision.

Suggested Implementation:
Style Selection Interface: Users could browse Suno song styles and choose one to start a new track. This would help them maintain a consistent direction without extensive prompts.
Controlled Customization: Allow users to customize their tracks by selecting style variations, ensuring a personalized but consistent result.

Potential Benefits:
Enhanced Creative Control: This feature would let users directly control the AI's output, matching it to their preferred styles.
Increased User Engagement: Reproducing favorite styles could strengthen user connection to the platform and support diverse creative expressions.

Thank you for considering this feature! It could make Suno more rewarding for creators.