Same request. It works for me when I specify lyrics and explicitly state "female vocals" in the genre.. but for other use cases where I want a male/female duet, and tag the lyrics for female vocals in verse 1, male vocals in verse 2, and both in verse 3, those tags are ignored and credits wasted. How about also refunding full or partial credits for rejected generations?
#š”āsuggestions-web
1 messages Ā· Page 23 of 1
š a shuffle function for the playlist would be nice
Can't tell if public song is owned by suno (song made on free plan), or owned by user (song made on paid plan). Would probably be good to know if you wanted to contact the owner to use it.
i think you can make explicit songs, right? You just canāt share them here and Suno wonāt promote them for listening.
I think i would request a āSafe Listeningā mode and the ability to turn it off. That would protect sensitive people and allow louts like me to listen to whatever pops up.
Great suggestion - saves cycles and server space.
I searched and didnāt find a similar report. The profile page on iPad Pro 13ā looks bad in Safari.
The songs are in two columns unlike the Home page which are in three columns.
The 2000s called and want their ugly right sidebar back. Make the right sidebar hidden until the lyrics are needed.
The space between the center and the left sidebar is 1/2 an inch from the end of the ā5ā on āWhatās New?ā and the āDā of the song (I measured with a ruler).
Using desktop version or mobile version in Safari doesnāt change the layout.
The screen resolution for iPad Pro 13ā is 2064 x 2752. The frames structure makes the site feel very cramped.
i just joined, so someone has probably said this before, or you may be working on it. But i think you should have something where you can change the album photo on your playlist (free). Thanks if you do!
a song being public doesnt change its ownership. if someone made it on a paid plan, they own it, regardless of whether theyve made the track public
The Library page looks okay.
The create page has the problem where the center frame doesnāt side scroll any longer which cuts the lyrics on the right side off.
This makes it impossible to create stuff on the iPad from the browser now if referencing lyrics from other parts when extending.
Also the Suno app is not iPadOS enabled so it is a bad user experience. That is why I am using the Safari browser.
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
not sure if this has been suggested, but would like to have a embed feature, code to embed it to a blog for example using trough a share button
Can we change the thumbs up and down to have one for the music and one for the vocals? It would be nice to find one that you like and be able to develop it more. So if we like the vocalist but the music isnt right we can keep the vocals and regenerate the music part, or vice versa
add " Prominent " to what you see as the most important, it should fix it 9/10 times (dont over Prominent prompt on everything)
example [Prominent Male Vocal]
Hey guys. I'm glad with the amount of credits for subscription. I just think that they should differ for a 2 min song, for a 4 min song, for a 2 min extension, for a 20 sec extension, etc.
I'd love to see it split into 30s blocks along with direct control over the max generation length (up to 4m or whatever they want the default to be). When I'm building up a song using extensions, I trash most of the generations before even listening to 30s.
... 'tis why my library is just 3.5 pages but my trash is 932 pages 
We need better sorting for the library. Only being able to sort by liked or public songs isn't going to cut it(and actually that is not sorting, it only shows them but it doesn't sort them). There must be at least descending and ascending sorting for generation day, name, times liked.
We also need better "tagging". Now there is 'liked' and 'public' but we absolutely need tag for full songs, different parts of song (part2, part3 and so on), versions (v3 and v3.5 for example). And these all must be sortable.
These all need to apply to search too, at the moment you are not allowed to decide what you're trying to search. For example, there is no way to search your full songs only. Trying to do that, you will get pages and pages nonsense.
Does this channel also cover suggestions for mobile browsers?
I guess it does then. š
sweet
I checked my iPad Mini and it displays with four columns. The create page pops up from the bottom. However, I could not side scroll to see the lyrics that normally would have shown up on the right in the create page.
For some reason I like the look and feel of the previous site layout and theme, but anyway, one thing I wanted to add here for suggestions is, that we need a option to have libraries for music and flexibility with moving music to different libraries, such asā¦ā¦ We have a main work area, for making, generating music and editing etc, then an option to move completed songs to a completed songs library, that way they donāt just pile up in the work area space, and we can put finished songs away in their own library instead of just the trash bin, then of course have the trash bin for songs that didnāt turn out good, that will be deleted. š
Would like to be able to transfer completed songs between accounts for organization purposes. As credit demands vary songs may ended completed on secondary account of via versa it would be nice to be able to keep all desired genres/artist/sounds etc on the preferred account wiout having to recreate a song to transfer it.
On the Suno channel on Discord, I almost got to know someone with whom it would have been good to collaborate if that person were more tolerant. This led me to a new idea about how Suno could strengthen its community directly on its main page in the future. Users could indicate on their profile if they are open to collaboration and specify in more detail in what way. The point is that not everyone can combine music and lyrics when creating songs; some do only lyrics, some only music. But through collaboration, it would be possible to find a partner who could complement what the other lacks... Of course, only if they like each other's work. However, this would require a new challenge, which Suno could consider now. If it were for commercial use, it is likely that both creators would be interested in clearly stating who owns the copyrights to the lyrics and who to the music.
It would be beneficial to have the ability to set a specific time range for a song, especially between 0 to 3 minutes. This feature could be useful for creating short music clips for intros, commercials, and other purposes.
I have a lot of finished song versions that I don't want to publish, but I'd like to take some fragments from them in the future. So I'd like to have an area for such things, too.
It do be like that, and it would be nice to have a delete all button in trash to remove all the songs you donāt want all at once.
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
/chirp
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
I find the censorship on suno front page, which is anti heavy metal to be very frustrating. Even with vague lyrics using heavy metaphors and mild pg13 horror covers, songs still get filtered. I work to hard on my songs to get no exposure outside of the occasional click in discord. How am I supposed to enjoy creating the music I love in this environment which is supposed to be open; where anyone can make great music, if only 13 yr olds can hear it. This is what heavy metal is, tell me how I can make this without feeling like heavy metal was just created by Satan.
Trash/Recover/Sort nav button integration # š”| Web version: Can we get a lock/unlock button on the created tracks nav bar? And is it possible to add the recover nav option if you accidentally delete, or need to do a quick recover back to the library added to the Details bar? Last suggestion for the trash, sort by date of create option. Thank you
I would like to make two suggestions which I think are something we've already touched on at least once in the past:
-
Add a "Character Voice". This would be premade voices for various genres that people can choose. By choosing a voice you get better control over the voice and AI won't go nuts making it sound terrible.
-
Add our own voices as character voices. Allow AI to take sample audio files from us to generate a completed voice.
I wanted to make a suggestion if you could put autotune on so we can record ficaria muito melhor porque eu canto alguns raps isso seria muito bom para ajudar na voz
Imo it is Sunos job to protect the creations of their users.
-
A song should be able to both be public, and only downloadable by its creator. No one should be able to look at the html code with their browser to pull this download link either.
-
Suno needs to give us a way to show we created a song. Your website, with date of song creation is not valid proof of ownership. Since suno does not provide this, it is easily possible to steal anyones song, even if its already uploaded to youtube, and claim it as yours. If say on youtube you send a copyright strike, they can counter your strike, requiring you to provide proof, which you cant. Meaning you now need to sue someone who stole your stuff with no valid proof you created the work to begin with. Its even easier if you upload their songs to a distributor, especially a foreign distributor compared to where the song creator lives, as they will likely not even know who to contact to have the song pulled down.
-
Suno should add BMI and ASCAP id's. Members of one of these organizations, like myself, can add our id to our songs on suno, as well as upload our song to BMI and ASCAP to at a minimum show legally we are the song writer.
They technically already have that by allowing you to download the song so you can add it to your distributor and publishing administration service. Only those services have the fingerprinting capability.
I agree that only the creator should have the ability to download. Also, there shouldnāt be the āextendā button on songs that are public that show up when it is added to a playlist.
My recommendation would be to make the WAV masters a bit louder so that they are as loud as the songs on Spotify. I've noticed that they are always a bit quieter, which is noticeable when you look at the WAV file.
You should import your .wav into a DAW and master it to the master specs that is listed on Spotify for Artists page to get the correct loudness.
Yes, that's what I usually do, but it would also be good if it came out a bit louder. Maybe it's just a suggestion š
How tricky would it be to add a filter for Instrumental and Lyrics songs? That'd be particularly useful for both the Library and Profile so I'd love to see that and we can listen through either of those choices
can the ai strictly follow the lyrics??? its keeps being annoying.....
Would it be possible to add the like button for a song into a discord video link?
For the next version I want MIDI instruments prompt support. Possibly gameboy, SNES, SEGA Megadrive, etc, but that would probably be legally impossible.
It looks like most times it can't. It just loves to repeat the same fragments or completely mash up different lines.
When I have problems that Suno went/run too slow on my Firefox browser, I got to situation where I clicked somewhere by incident (this happens when the reactions of the loading page are too low like 10 seconds later and so on) and this triggered extension of random a song in a list of songs. I think that in these cases where the Suno is very slow like more than 1 second reactions after a click... there should be something like "stop" button, to stop immediately the process. I feel it unfair when I as a free user having like 40 credits at that moment, suddenly I lost another 10. And with 30 credits that day, I cannot do "anything". I mean one song cannot be finished with 30 credits (with 40 credit maybe yes, maybe no). I know that in other browsers the desktop app runs better, faster, but I don't like the other so much like this one.
@mental horizon I'm making a youtube music video and I'm wondering how I make a ai video from audio peice I had created. Could you please help me?
I just add an image that's shown throughout the whole song.
Oh thanks
I'd really like Suno to permit us to use our own voices. I made a short excerpt with my voice, and it uses it very well, but I can't make it public.
Dude... don't ping the devs for common stuff the community can help with.
Sorry
Not the end of the world, but if one person does it, monkey see, monkey do ā¤ļø š
I haven't looked into it, but I see a LOT of people using LumaLabs or Hedra for this kind of stuff š
Oh ok thanks
@tepid trench how did you make that terrifying furby video?
What are your capabilities? Win\Lin\Mac\Android\iOS?
Are you comfortable working everything in the browser?
Windows in Google
Windows computer with Chrome browser? Ok...
What kind of video are you looking to make? A simple repeating one or something with a bit of a story?
One that follows the song so a story
OK, understand right out of the gate that kind of video is going to take time and effort. It may even cost money depending on how much time you're willing to put in on free tools.
How much of the video do you want to have be fully AI?
All of it
So something like this: https://www.youtube.com/watch?v=5A3Do1jRt6o
⨠Step into a nostalgic journey with "Plaisir d'amour," a captivating Retro Synthwave track featuring the iconic sounds of the TR909 and Roland Juno X. This mesmerizing blend of classic synthwave elements and modern production will transport you back to the golden era of electronic music, evoking feelings of love, passion, and timeless memories....
That was made using Midjourney (an AI image generator) to create the images, then I used PixVerse (a text to video and image to video AI generator) to animate the images I created. Finally I used Hedra (a talking portrait AI generator) to make the characters sing.
Oh
I'll try that
But I have this so far
It's mine
That seems like a static image?
What is your story? Who is your charatcer? Is your character.....actually I apologize. We should move this to offtopic.
Alright
Suggestion: Medley mode for playlists with max song length slider or menu so you can review the content of playlists more quickly, maybe a 'skip intro' toggle to forward the song to some bass drop or hook . Example 279 minute playlist https://suno.com/playlist/1537a7f8-26da-47e6-9374-50641a453ed0
Any chance of a queue/stream/radio system? I understand it eats up a lot of credits, but I like to set up a prompt and then just have constant new songs without having to manually generate them.
Ended up creating a wrapper around Suno endpoints to support this but wondering if there was any official plan for this type of feature.
Essentially just a prompt radio station. Works well with instrumentals. Especially if you find a prompt that fits your tastes.
Suggestion : User Follow/Following indicator on any main song display.
Can we fix the edit displayed lyrics box so that doing literally anything doesn't close the text box without saving all the edits? this is legit infuriating.
Suggestion: Sort and Shuffle playlists, seeing the most recent songs at the top of the list when refreshing the playlist page would save a lot of scrolling to the bottom of it
Even clicking INSIDE the text box, but slightly left to the text (like youre trying to select from the beginning of the line) closes the text box.
Here's the crummy code for running this locally if anyone else is wanting to have this capability: https://github.com/liqMix/sunoStream
Hi I suggestion that I don't has been shared, 1, to beable to cancel generations yes we use credits maybe small refund when it's canceled. Not only will this free up resources on the servers and back end . Because I don't listen to full generations when I playing with my lyrics . I don't need a video generated every generation, may have it on demand.
Okay, first of all... LOVE SUNO, I have to admit, I'm officially addicted. I do have some questions and feedback though...
First of all, where is the "Empty Trash" button? I painfully managed to put all my junk files one by one in the Trash, I'm not deleting them one by one from the trash, that would near to impossible. I think no one does. This means servers are getting needlessly full. Also Bulk action would be nice. And please lock public files for deletion...
And second, it would be great to be able to extend in the front or in the middle of the song as well. And an embed code for your playlist would be great.
@robust jewel thank you for that
yo idk if this is a bad suggestion but
can we get an option for the ai to experiment more with stuff like..... examples of samples being used in songs and use em more creatively than just copying the beginning 5 times
Iām trying to make court music reminiscent of Japanās imperial court, heavily utilizing the sho instrument (a free reed instrument similar to the Laotian khaen and the Chinese sheng). However no amount of prompting has really been able to produce that specific sound that Iām trying to create. I wonder if Suno being able to more heavily prioritize specified instruments would be a nice feature to explore, or if these rarer instruments need to be manually tagged somehow
Not the best shortcut one could expect, but tab-tab-tab-enter works.
Or alternatively, shift-tab, shift-tab, enter
Still better than having to use mouse, imo.
One could also use autohotkeys, but... I won't do that š
on the create tab, after create the music, list the instruments used, so we can reuse prompt and remove some instruments of the list and add others
Hey guys. A quick suggestion / feedback on accessibility of your website for screen reader users. I see you are doing some work already around the site, but the most important functionality, such as creating songs, playing the ones already created, etc, are still lacking. Could you please consider improving these areas? I think the dropdown menus are not accessible with keyboard, etc, e.g. selecting genres. Let me know if I am not making sense or missing something, of course! Thanks for giving us so much fun!
Within "Create" it would be useful if there was a little "+" button that you could quickly add a song you like to a playlist (or a WIP folder of your choosing, or an automatically generated folder with the same title as the song)- I currently like songs to sort them, but its a really messy way of doing it and not useful. I get there are playlists already, but its a bit fiddly to organise songs into playlists. Needs to be easier!
I have a suggestion that wecould click edit lyrics and change the lyrics in a song but the music and voice would stay the same. It would be like ai inpainting.
I just would love simple edit capabilities to be able to cut off the endings after the ending. However, what would be amazing is being able to trim, then arrange, then have the AI process into a whole song so you can get exactly the pieces. you want all together in a song.
Suggestion: Enhance the AI to compose songs with classical themes and patriotic melodies, including national anthems.
It is highly beneficial for authors of royal novels to have the capability to compose patriotic songs.
Suno should already be able to do this, just not directly via entering "Patriotic" in the style box. Think about what musical styles might be used to define patriotic music... maybe something like: "March, Anthem, Hymn, Folk, Classical, Orchestral, Brass Band, Military Band, Choral". Breaking it down into styles like that will allow you to further customize it to your own preferences.
It would be amazing if we could train our own voice models for suno, it would give a lot of creativity and new ideas to us, and also explore all the potential of suno, obviously with moderation! I would love to see this in the future
Have you tried using the upload feature? If you upload a sample of yourself singing/vocalizing and extend from that, Suno will often create an approximation of your voice.
Yes! Itās amazing the ability that it has to make something similar to the voice you give, but everytime itās different, so it would be nice to have a consistent voice
Plus, right now you can't make public any tracks made with a voice you uploaded. Which is very frustrating.
Yea, that's frustrating
A paintin function to recreate parts of existing songs. Something like the "Reuse Prompt" for the voice "Reuse Voice" so that the new created song have the same voice like the old song
is there any way to sort the library search by when the songs were generated? now it's all scattered around with random dates making it hard to find the exact one when several songs are named the same thing
- Stop suno from auto scrolling down after generations
- Option to stop pop ups
- The pop up for too many generations comes over the play/ pause etc buttons - it needs moving or ideally an option to stop those pop ups
- Infinite scroll is desperately needed on create/ library
I'd like to have an option to move tracks into playlists, so that they are not displayed in the list of all created tracks. Right now they are just copied there.
Having AI use people's voices to sing in other languages would be good.
The style of music and Title box need to be significantly bigger
I got good results by typing 'war beat, war motivation' under style (the song's genre description in Wiki) which updated this old song I liked from 1917. https://suno.com/song/2e245699-1511-460f-a938-757ed30d168d
war beat, war motivation song. Listen and make your own with Suno.
Ow, and a trim function to cut your song on a desired time-code, would be really great!!
I'd like to have options to add fade-in and fade-out to the beginning and to the end of a track, respectively.
i've got a suggestion, get suno able to understand IPA
Make a "shazam" for type of music
Send all dislikes to trash instantly. In fact delete them from my account entirely. Get rid of trash can. Whose looking in there, and why store infinitely increasing garbage. Not good for servers long term.
I'd like to suggest that playlists have numbers so it is easier to scroll through to find a song at a specific numbering.
"16-bit video game" and "chiptune" are not the same thing! And 8-bit and 16-bit are also distinctly different.
Suggestion: when producing music have it split out the tracks (e.g. drums, vocals, etc) so that you can download them as a whole song or as the parts to make it easier to edit songs yourself afterwards or remove vocals to add your own
(Bonus points if it can save them as MIDI)
I'm a little bit questioning whether it's a bit unfair... A bunch of scary pictures can be on the trending list. Normal people like us can't even get through it. Even the pure music that I just sang by myself is no more. Isn't it a bit special treatment?
Doesnāt he even count the pictures = = Many of those are simply PG-13 on the list.
This genuinely confused me, because I have no idea why it would need to understand beers. Today, I learned that there is an International Phonetic Alphabet. No pale ales for the AI.
i didn't even know Indian Pale Ales existed
Not sure where you get the impression they're the same thing. Is this #1252257127326548028? Creating a playlist for 8-bit / chiptune / videogame music isn't meant to exclude songs that don't explicitly fall into those hyperspecific genre descriptions, -- nor could we be expected to create a separate channel for every possible subgenre.
If enough people seem interested in exclusively hearing "Sega" music instead of "NES" music, then creating a 16-bit, 32-bit, whatever-bit separate channel is a logical conclusion š
Perhaps if you'd like to see that included in the community soon, you could start a post about it on the discord-feedback link I shared above, or find others who have already expressed similar opinions in https://discord.com/channels/1069381916492562582/1253055412303958099?
I mean, you're 100% right that they aren't the same thing, but for the sake of discussion are you willing to at least concede that 8-bit Video Game and Chiptune are closer in similarity than 8-bit Video Game and EDM?
Trying to reorganize a playlist on mobile is a pain imo. Any help?
You know Youtube Live Chats Channels like Lofi Girl ? https://www.youtube.com/live/jfKfPfyJRdk?si=Zkh3vwZSxyzwwyVY If you could package something to set that up into the account it would be huge. Like if you can stream a suno playlist to youtube on a loop with a simple looped animation, that'd be hot!
š¼ | Listen on Spotify, Apple music and moreā https://fanlink.tv/lofigirl-musicš | Lofi Girl on all social mediaā https://fanlink.tv/lofigirl-socialš | L...
while in your trash, an empty trash option would be nice, with some filter options like "remove trash older(date) than (3 months/6 months/12 months)" and "remove trash shorter(song length) than (0:15/0:30/1:00)"
Suggestion: Showcase Discord Playlists, ie, somehow get Discord embeds into playlists.
Over two days I filled 52 lists, with some duplicating and triplicating of items to ensure tracks got more than one chance to be heard. Then I got carpal tunnel syndrome again.
Have to scroll 54 items down to add a song to a new playlist now. What was I thinking?
It was fun while it lasted and people seem to like having these lists that are easy to explore.
see my profile: https://suno.com/@resoundingcounterpoint638
I assume you have a Discord bot in Showcase channels that takes requests for suno tracks and adds it to the playlist of a Showcase Suno account that can be followed and easily linked to.
Why? Because the Discord is a community that should be recognized and fostered. It's a resource of infinite knowledge, know all there is to know.
/chirp
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
Suggestions:
- Playlist repeat issue (currently the playlist doesn't repeat, and if you try to repeat it'll only repeat one song).
- Pro users ability to download video without the suno watermark and without the "style" text under the song title.
- search needs to be improved -- it returns a lot of resuls that make no sense;
- a sorting option needs to be included and
- search and sorting needs to be added to trash as well .. it's a nightmare trying to find (accidentally) deleted files
There was already suggestion made on negative prompting when users want to exclude negative results like unwanted instruments in the artwork. But there is a lot of phenomenons which are hard to describe or to name them, or the naming and description to AI contains ambiguity which again results to unwanted results. These phenomenons are often created with certain conditions, which can help to identify the source of problem and improve the feature of negative prompting, if implemented by developers later.
To describe this problem I would use the noisy like metal sound which is often generated by synths or by drums, probably by hi-hats or cymbals. I have already identified the source of the sound which is very similar to REMO spring drum. Examples can be found on youtube. Perfect sound was generated from a aluminium can, which had a metal spring soldered to its bottom. The performer located himself in bathroom and with a echo he hit the spring with his finger. This created surprisingly very bad noise similar to the depressive or dark atmosphere in songs when negative words like divorce, illness, death of a child were present in lyrics. So here is the phenomenon of sadness expressed by AI using sounds similar to the REMO spring drum or just spring drum. If it was me and I would want to describe this problem having more complicated semantics, but clear for developers I would describe this negative prompt like this:
if (mood=sad or mood=dark) then exclude(instruments=[spring drum,synths],effects=[reverse,reverb,metal noise,remo spring drum effect])
If the mood is sad, then the specified instruments and effects must be forbidden.
Another solution using pair would be like this:
on mood=[sad, dark, melancholic] or mode=[harmonic minor scale,phrygian] exclude instruments=[spring drum,synths] and exclude effects=[reverse,reverb,metal noise,remo spring drum effect]
+Editor where user can add the conditions, moods or modes when the effects or instruments are forbidden
It would be glad to have an option not only to extend from the end but to add a fragment in the middle of a song. For example, add a guitar solo between 3:40-4:00, etc.
Some channels don't seem to show up for me until after I've clicked on them in a message like yours. I remember the same thing happened with #š³āmini-polls
Can Suno create sounds of nature? It looks like it can't, because I tried several times to start with sounds of thunder and pouring rain but it just ignores such prompts.
Have you muted the channels?
I've had some (fairly limited) success with v3. Not so much with v3.5
like with this one: https://suno.com/song/31a91d5a-1222-4a13-a1fb-6c3df8ca7ccd
Rain Storm, Thunder. Wind, Doom Metal, Horror, Ambient Noise song. Listen and make your own with Suno.
Can we expect to have an option to create one track instead of two at least for extensions? When I have 5 minutes of a song with 5 identical verses and choruses, I don't need two versions of virtually the same when I extend it.
Seemed that some were premuted. That, and Discord seem to be trialing a 'select which channels you follow' thing, with only some preselected. Or at least that's how it seems to be for me.
Yeah, I believe that a certain number of them are pre-followed, and some are opt-in -- that way some channels act as a lobby, while others can try to remain easily accessible, but at least 1 intentional click removed from people "accidentally wandering in".
https://discord.com/channels/1069381916492562582/1254418510885687366 There's an existing known-suggestion for this idea, if you wanted to upvote it for visibility? This way the devs know what's important to them, and what can wait.
I would love duets with male and female singers.
For Singers: It would be cool if there is an option with lyrical songs where you can get a fully sung song, and then if you wanted to sing the song for your own cover with the background musical audio but without the AI singer, instead of running it through a vocal remover, it would be nice it there was just a clean built in vocal remover.
add hardcore/hardstyle... songs please
instrumental ?
I understand that there is instrumental that doesn't have lyrics but if you do instrumental then you don't get lyrics and if for instance like me, you suck at coming up with lyrics that can be sung with the melody, you can get a full sung lyrical song, know how to sing it, then just sing it without the vocals... idk I thought it was an alright idea, maybe not.
ohh you mean if you have a good song with lyrics, want to remove it ?
yes good idea
Yeah like a full, complete song, with lyrics and the AI voice singing the song, but like have maybe a download feature for lyrical songs to just download the melody so you can sing to that song if you want to.
i still dont get how you can do all this, but cant just add another button after the songs done that says get full vocals, and have the ai run over it again and fix it up, i have so many good songs, but i dont want to just use your beat and instrumentals, i always get the stems then remake the song but its not up to radio standards like when i have actual artist sing, theres always artifact, hi hat and bass frequency mixed in my vocals, also some of the vocals frequency's go into the intrument stem. and i know its not just the ai vocals cause ive made some where the first verse, is so clean and mix and mastered perfectly to the point no ones noticing thats ai. but thats cause there was no kick and hardly any intruments playing she was just singing. then when i master your original songs they come out good. but i dont like just uploading your songs. i see it can be fun for people to make music that dont want to go through the hassle. but for people that actually want to make music or even samples for there songs theres a huge issue for us. everyone can get a daw, and do what they want but not everyone can sing, or have singers at all times for everything they want to produce. and thats where suno can come in heavy right into the industry. but untill you do what i said actuall music producers are screwed over by not being able to simply grab the raw vocals after the fact.
the issue here is that this isnt what Suno is made for. you cant get just the vocals raw from suno bcos it doesnt generate raw vocals but an entire song.
there are lots of AI generation tools you can use to sing (even terribly) and convert it into a realistic vocal (eg Audimee) or program in lyrics and write pitches in a piano roll and generate it in a realistic sung voice (eg ACE Studio), but Suno is made for generating full songs. I produce, write, sing, mix etc myself too and love using Suno as an inspiration point but it's not and isn't made to be a vocal generator.
You can upvote this however to show the devs that you're interested in it: https://discord.com/channels/1069381916492562582/1253172964728377455
What vocal remover you use...?
I try googling this but none of the tools I've tried worked very well for my songs...
/rap
This has most likely been mentioned before, but please let me both search for something and toggle to only show liked songs. Please.
Btw before you had the search feature I made my own ui with a search option using react and by simply saving new songs added to my own db (only the json data, Iām not ddosing u guys trying to download all audio files š ), and in that ui I simply download all the songs data to the browser and search and filter client side. With a bit over 10k generations it still is no match for a low end smart phone to handle filtering all of it client side.
Because of this I can filter and search with like 2ms response time and no server load. If you did something similar it would probably help the servers drastically imho.
Querying server side makes sense ofc when it is searching all users tracks, but when just searching your own tracks it doesnāt make sense at all not doing it client side.
Just a lil tip from one dev to another.
Edit: I fetch the data by fetching it from the songs list in the html whenever I visit the site, so Iām not doing requests to fetch this. Just to be clear that I am not doing any extra spam to the servers.
Suggestion -
dont know if this has been brought up yet..
An advanced feature to force certain aspects of generation, such as a toggle to force the song to end.
i have found with 3.5 it really doesnt want to end songs, even when using specific prompts to try force it. ive found myself using v2 to end songs because of the shorter clip length, but the sound quality doesnt seem as good
also, im sure this is brought up once a day, but being able to export stems, or even a suno created stem extraction tool would really push this app/website to the next level
This is a billing model that MANY other services have adopted; I don't see it changing any time soon. It also affects the monthly credits for paid accounts - use 'em, or lose 'em. š¤·
If u get over 100 plays on a song, 10 credits gets added to your Suno account automatically
suggestion: more flexibility on the timestamp for the extensions. cause sometimes i'm force to continue at a time where it's litteraly not in tempo with te music. so not just only mm/ss, but mm:ss:millisecondes (don't know the abreviation fi there is one, sorry)
Just a tip in the meanwhile; if you extend a song and then extend it again only like 2 sec into the previous extension and write something like [End] you can usually get it to end because the context gets cut off because of how little audio there is to build on from the previous extension. Kind of a waste of credits tho, but yea. Also switch to v3 to end a song, v3.5 is like that guitarist who just canāt ever, ever stop playing⦠no matter how many stop signals you try to give him
And sometimes this guitarist stops when you don't want him to.
it's hard to get a guitarist to shut up... 
Again, please display somewhere the total number of pages for our library.
The best would be something like 5/89 where 5 is the current page and 89 the total number. But i guess that this is going to take some place. So display the total number where you want . We should not need to wait for a big update for this minor thing.
To be honest, I only have tried a free vocal remover online called, Vocal Remover, but I'm sure you already know, it didn't work that well. Hence the idea I had of adding that feature built in.
The potential issue with Suno offering it would be the same issue
Suno doesn't have sperate vocal/instrumental tracks
I'm sure it can probably be done better internally, but fundamentally the same issue. Have to separate them after generation
It is in the poll though
Devs are certainly aware it is a much requested feature
Moises is pretty decent at splitting audio into multiple tracks. Itās an app and has like 5 splices per month for free if they havenāt changed it. Or a cheap subscription for unlimited splits. Also lalala.ai is sometimes better.
Awesome thanks, I'll check it out sometime.
The recurrent issue with all these tools is the output quality. When you remove vocals from the input audio you remove also some frequencies from the instrumental tracks. It results to poor audio quality
those are all trash compared to suno, i want to give my money to suno. im sure they can program some stem separation thing, then have the ai run over the vocals to clean up frequency's. why allow other company's to take some of your profit when you can easily destroy them.
I've had a bit better luck with a tool called wavepad
We need to have an undo button
So I'm loving the ability to edit the displayed lyrics and get rid of all the Meta Tags for a cleaner look. But I think it is imperative that the original, unedited lyrics be saved with all the Meta Tags, as well as a "Lyrics Edited on DATE" on the songs page.
The reason being that the US Supreme Court ruled that 100% AI generated content was not eligible for copyright, specifically stating "human input" is required to be eligible for copyright protections. They didn't specify exactly how much human input is required, and I believe it will be a matter of future litigation.
The raw, unedited lyrics would be crucial to demonstrating human input in the creation of our beats.
@sonic fern Also maybe original lyrics vs generated lyrics would be nice.
Would love to be able to get instrumentals of some tracks(during creation as double of the credits or after the fact) without having to use other tools to do so and making the audio wore in the process.
How about a place just to generate lyrics and show the original prompt so if we don't like the lyrics we can try again.
I have a (probably extremely difficult suggestion) I'd like to see harmonic unison singing. Like you get in Broadway musicals. Two people singing the same or nearly the same line 3rds or 5ths or whatever apart at the same time.
Have Song Descriptions been considered? It'd be a nice QOL feature, which will allow people to write descriptions for their Suno songs akin to playlist descriptions.
How about a notebook or notepad of sorts where I can store lyrics and song styles and things while working on a song so if the page refreshes or playlists crashes I have my stuff still.
I get around this right now by having multiple Suno pages up in my browser where I 'keep' this sort of thing up, but would be nice to have just the one tab.
A feature to find what playlists your songs are in would be cool. Knowing how others classify your tracks and what was liked about it is helpful for inspiration.
And as to original lyrics, it would be nice to have an option to indicate that you wrote them.
Would it be possible to add another extend function but for the beginning of the song, so it's basically creating an intro?
Would it be able to add a feature where you transition from one song to another seamlessly, so it's like one whole song?
This would probally not be allowed for privacy reasons pr copyright idk but:
A way u can upload your own voice models where the ai sings with. (Pth files or onnx files for example)
So uh... I've got some alright skills at playing various instruments, and have played around with image generation AI before (mostly Stable Diffusion, where I'd often combine my existing 2D and 3D art skills with generated stuff to get the results I wanted). Now I'm trying to figure out a nice workflow to use Suno to either make music that I can be proud of, and/or improve my existing guitar and singing skills.
So far, I already had a few neat moments by dissecting a few particularly nice Suno gens using Fadr's stemming tool. Learned a few things about song composition and found inspiration to mess around on my guitar. I'm not having as much luck trying to actually work with Suno's gens, though. Compared to the tools I usually work with, Suno doesn't offer much control.
So here's a few suggestions that might make raise the skill ceiling of this tool:
-
Allow uploading a song as purely a reference (not extending, just feeding the song into a cross-attention layer in your AI so it influences new songs). This would hopefully allow users to steer music in specific directions without needing to use the first seconds of an existing song. I don't know if the AI actually "sees" the parts of an existing song after the point that it should get extended; if it does, then this idea would basically just mean extending from second 0. I guess I could try that one. Ideally, the influence of an existing song should be something the user can tweak.
-
Allow remixing a specific part of a song (timestamp X to timestamp Y), with the same influence strength option to determine how much the song should be changed. As long as the BPM stays roughly the same, it should be possible to insert the newly generated part into the existing song, though it might be tricky to get the timing right. Feeding the original part into a cross-attention layer should allow the AI to figure out the timing on its own, though, I hope...
- Generate different parts of the song separately, so they can be remastered or changed more easily. This one could be a bit tricky: basically the rest of the tracks for a song have to already be available for any new track, so where do you start? Well... maybe with a song generated using a simpler AI, and then you generate each separate track on top until the original track is fully replaced. I feel like this could also result in higher quality audio for each track, though it would be harder to gather a good dataset with separated tracks. Don't know if that's really possible. You could possibly use something similar to Fadr's stemming tool to separate tracks for your existing dataset, resulting in a somewhat low quality dataset that'll at least teach your AI what a separate track is supposed to sound like, and then you could train your AI with a loss function that just checks whether the combined tracks match the expected output for your dataset.
Alternatively, maybe you could teach it to also generate decent tracks without any input to work with, in addition to generating tracks that work well on top of an input?
Having separate tracks would also allow the user to change the general sound of any particular track more easily.
I don't think we're too far off from #1. You can already kinda do that (with mixed results) if you leave a few seconds of silence at the end of the upload and extend from there. Suno often seems to treat that as a reference. You'll have to trim that reference off with an external tool after the fact if you use Get Whole Song on it, tho. Annoying, but it does work. These all pop up quite a bit as suggestions (using uploads as a reference, excluding uploads from Get Whole Song, and trimming tracks)... So they're probably somewhere in the back of the devs' minds.
Inpainting is coming with v4, which may address #2; fingers crossed.
Re: #3, stems are a frequent suggestion. It'll likely be via a tool/function that separates them after-the-fact since Suno itself does not generate the songs as separate instruments/vocals.
Oh, that's clever! I actually considered doing something like it, just a bit of reference audio and then silence, but didn't try it yet.
Inpainting sounds pretty exciting, too! I guess I could've just called it that instead of describing it.
Here's an example of Suno using an upload as a reference from horrible playing into a phone perched on the corner of an end-table: https://suno.com/song/70a9eccf-1652-4e99-959c-d8e9a0f356b8 It captured some of the character of the guitar as well as some of the noodling
Pretty interesting
Can't really tell where your bit ends and the AI bit starts, which is good, I guess.
heh. My bit isn't in there at all. The extension came out under 2 mins, so I didn't merge it back in. 
you can use "Get Whole Song" on it if you really want to hear it, but it's... not good. lol
You can use uploads for setting the tempo of a song, too. Just tap out a beat with your fingers, or use a metronome.
After using suno for sometime I noticed there is a default pattern it has, I love the program. It would be great to pick your own AI vocalist for a specific track, possibly even preview before making a comit to the song. What I'd love to see, is possible verse extensions from 4 bars to 8 or 16 bare measure, and an option, to have AI on the chorus only. Sometimes the Ai will take the chorus and add it in the verse section. Maybe a new prompt such as [Empty Verse 8 bars or 16 bars] what ever the user would want. Would be very nice. It would unlock the potential of never ending possibilities.
would be nice to customize song structure ie 4 bar intro 8 bar chorus 12 bar verse etc
Can we get a built-in trimmer/cutter of the results?
Often I would get an inferior bonus track
please i need shuffle button!!! š
@muted delta@tall cradle@grave salmon@carmine flint@open rock@solar sable@timid hatch@echo canopy what about a wonderful Shuffle Button to randomize sequence for a playlist or library?? thanks!
https://suno.com/song/8554a12f-0f21-40d8-bf48-1a708d571aba i ask also on that way
Shuffle, Pop, Rock, Driving, Electronic, upbeat, catchy, beat, guitar, bass, techno song. Listen and make your own with Suno.
Yes I agree on this!
I would like to have an option to combine a whole song, for example, without Part 1 if it was an uploaded audio. Because this first part is quite often not a part of the combined song.
The number one enhancement I would like to see is reliably ending songs when they should be ended, rather than trying to fill space.
an idea that had a button called "Add style" this add style you could upload a maximum of 2 minutes of a song and choose a keyword and this keyword could be used in the prompt to try to copy the style of the song
Make a Station that randomly chooses songs. You could add 10 second commercial spots and pay people small sums or reward them in Sunobucks/coin(tm) if their songs get played....
hi, can you add a "notes" field to every son, that s only accessible to the creator. It would be great to have during composing songs.
PLEASE fix the bug of slecting WAV to download and the audio doesnt appear or it appears 5 min later to get downloaded!
My suggestion is implementing something like the recent paper
"MusiConGen: Rhythm and Chord Control for Transformer-Based Text-to-Music Generation"
I don't think suno has precision control of the rhythm and chords of a song.
chirp create song from text
Chirps deprecated and my suggestion is simply implementing better chord and rhythm control maybe Inpainting in V4 will solve all this.
Hi, sry if this has been suggested, but a ultima tear of credits. Iv spent 90$ this month . And would love a bit of bang for my buck
I think being able to start a "Full Song" from a later generation would be a nice feature. I had a song I was finishing hallucinate the beginning of another in the final gen when I was trying to end on silence, and I ended up turning that into a completely different song. Since it's entirely contained from start-finish in that initial hallucination gen, being able to split off a Full Song from that start point so I don't have an 8-11 minute song that contains the original song + the new one would be really nice
After testing the new function, the effect of separating vocals and music needs to be further enhanced.
Especially in the musical accompaniment part, there is still obvious residue in the vocals. I hope this part can be improved. Thanks for the conclusion of the above test.
R&B kayÅ, chill Spoken Word female vocals,Groovy,HI-Q,Slow,Soul House,Synth House,KPOP,Heavy Bass,R&B,Soulful song. Listen and make your own with Suno.
This is the result of the test. The human voice was not used for demonstration.
The remnants of his vocals will become more and more obvious
Based on my own experience in the recording studio, I put the music and vocals on the left channel and the right channel respectively... This is the method that recording engineers used to do in the past. If it can be implemented in this way, the effect will be good. It is very convenient to branch out ( Of course I learned this a long time ago at a record label)
Suggestion: Add (BPM) somewhere in the track title so we know exactly and not guess or test! Thanks š
I never ever want the video the songs generate, only the audio.
I don't think I'd be alone here, so it's probably not a bad idea to add a checkbox to opt out of the video generation.
seems like it would save you a lot of processing power for how many generations you must be making that may be going to waste.
SUGGESTION: Flag 'Upload' and 'Lyrics Edited' on songs.
With the new super cool features Edit Lyrics and Upload it's easy to be misled by prompts you see in Trending songs. A little icon to flag songs as Upload or Lyrics Edited could be nice.
add a button repair that keep the flow of the music / voice and remake the music without bug
video one ia / music one ia
they don't waste anything š
what does this mean
they use differents ia to generate the music and the video ^^ its mean, the music ia don't loose anything š
@tacit idol im true ?
oh right. that doesn't matter, it's still wasted server load. I didn't mean that the music quality would be affected
ohhh ok sorry this is what i understand ^^'
The music and the video are generated simultaneously.
ohh ok
There are a lot of suggestions for inpainting, clip uploads etc. Here is my take after spending about 5000 credits.
- Sliders. Yes, sliders. Imagine having sliders for tempo and rythm. Awesome.
- Check boxes for musical scales. Want to have it in A-minor instead of G-major? Just check the boxes.
- Check boxes for voice modulation. Want to hear female growls instead of male operatic? Just check the boxes.
- Check boxes for instruments. Want strings instead of flutes? Just check the boxes.
I admit this is an utopia, but we gotta start dreaming somewhere.
To start with, I believe the software should be trained on music fundamentals. Train it on metronomes in different beats and tempos. Train it on scales repeated on different instruments. Train it on chords.
The software is amazing and a big boost to my creativity. But I lack the control of just putting my 20 year old synthesizer to play a scale that I can start composing over.
I've spent a lot of time in the image AI world the last year. Models, Sampling methods, Seeds, CFG... Such alternatives would be great to play around with in Suno.
Imagine generating until you find the "right song", then locking the seed and tweaking instruments and tempo.
Pleaase, buttons to sort files and playlists!
For the next algo version, minimise false, long finishes and bonus tracks.
We need an option to reuse the same vocals of a song we just generated for another one, or give us a library with voices to use, or let us make our own as well! That would be so fun
The progress bar must be more dynamic and refresh into the finished song length.
did i just predict the future?
A quick fade in/out option
Often the extended generated track ends in the thick of it.
Trying to understand how Stems can be used. There should be an option to add a custom vocal file to an instrumental stem.
Better soundquality is most needed
This new update is a step in the right direction and it clearly needs more work, im sure it will improve over time but what i suggest ( if possible technically) is to make it so when Suno AI generates music it does so in separate stem tracks/channels ( just like in any DAW) like Vocal channel, Bass channel, Synths, Drums etc. So many tracks get generated that have some parts that are absolutely amazing and it would be awesome to re use them in another track, and please add tempo tracker AND what key the track is in. Take a look at ( Mixed in Key ) system they use and perhaps train your AI with it.
This would be really cool!
thanks for the latest update! As suggestions for F from the mini-poll I would really like to have a better audio quality like 256 or 320 and maybe the gain could be louder.
Besides that thank you very much for Suno. I really enjoy it ā¤ļø
Can we nuke the piano from space? Iām tired of Suno adding piano to orchestral music, very rarely does an orchestra have a piano. This seems like a training problem, everything always turns into a piano.
This š¤
PLEASE fix the bug of slecting WAV to download and the audio doesnt appear or it appears 5 min later to get downloaded!
can you make it possible to Edit an generated song,
e.g. keep the complete feel of the song, but be able to adjust the lyrics to either remedy simple typos or adjust them to fit better in the current sound feel.
Just having a negative prompt and the ability to set weights for specific words, like in A1111 (for AI image generation), would be huge... and pretty easy to implement. Then you could just say "anything but piano" by adding "piano" to the negative prompt.
I have a feeling this has been suggested before but wouldn't it be nicer if we could remove commenting on suno-showcase channel. It is frustrating when some ppl just use the space as their own chatroom. just leave reactions but remove commenting.
can you make it so that we can upload multiple stems of a song and then create a whole new song around those stems? would be EARTH SHATTERING
iām not sure if this will be possible but some way of changing the genre of a song you generate/upload would be amazing. For example you upload a rock song and tell the AI to āCreate a piano rendition of this songā and it recreates the song using only the given instrument and it makes changes to fit the vibe more
i can still hear vocals in my stemps
This should be possible a genre variant of Inpainting or rather a music 2 music genre switch similar to how image 2 image works
Is there a way to download the LYRICS like SRT file? alights text to time
i suggest when new feature come give a week for free user to try it.
I have a suggestion for Suno, keep doing what you're doing because this STEM update is everything I've wanted and so much more. Shout out to you all for constantly improving my love for music ā¤ļø
Hi is it possible to suggest, gifting credits to people or gift subscriptions.
Some talented people out there that are budget friendly
SUGGESTION: User comments on Suno songs, but not text comments, comments must also be Suno songs. (Even if text comments were allowed one day, allowing this alternative modality could be really fun.)
I already have my comment ready then:
https://suno.com/song/57ae29a3-51ae-42c6-b9ec-b9fcfcc9cf4d
80s, new wave punk wave, female power, post-post-vibe cassette song. Listen and make your own with Suno.
eyyyy suno team, great change on the front page ui, waaaay easier to navigate
The new home page is worse than the old one
it seems to scroll very slow (on highspeed cable connect) and favors people who have external followings for their music you need to get like 10k plays to hit frontpage now. Bad for niche/sub-genre creators
Hello, I want to suggest team to add export .srt file to easy make a subtitle for videos instead of export video, because effects on video is not met requirement for social video
Sometimes i want only start one song job and not two. Have a first look if suno can pronounce my text or that i have to change something. A option to say how many jobs suno should start would be nice.
What does the new home page look like? I don't see the difference compared to the old one.
there are so many thinks that should be fixed IMHO:
- no dislike button
- no reuse prompt of someone else
- no extend for a song of someone else
- block copy lyrics of someone eslse
- a shuffle button for playlists
- an android client
š
bring back the "show all" button, at least on desktop
Not a Fan of the new Scrolling... atleast for everything.
Its also not stable, it makes it worse over time and really slow.
Also it kinda discourages you to go to the last Songs, since its annoying to scroll sideway
for me the explore page would make more sense if i could atleast use a search function. Eg when i type in pop, it reduces all the styles to the styles with pop in the name...
Suggestion: Make the trash folder searchable.
how can i replace vocals with instruments, for example if i hum a certain melody then it takes my humming and transform into synth or guitar (in my case it just generated my humming all over again), i believe the next updates should about giving the user more control on many aspects, also if they add a button where it says inspire it is like extend but generates a song that is inspired by what you upload or what is in suno
Yes, Extend from a personal file should have at least several options: extend directly from its melody or vocals, incorporate its melody or vocals, make something inspired...
...and change the sort order so that the most recently deleted files appear first, 'cause having to do this with a veeery large trash bin to get to them is just silly:
I donāt wanna hear my humming cause my voice is bad but i want it to take the melody i hummed and make it into the style i want exactly
And there should be a slider where if i want the song be influenced by the instruments alone or the melodies only or both, imagine the beats that can be created
Yeah, sliders, we need sliders. For everything.
I hope we get news if they are considering adding it in the near future
Hate to be this guy but, every new change on the browser lacks even the minimum optimization. Im running suno with an old laptop, it was perfect then overall ui update kinda mess with everything and it was a bit laggy but still usable, now everytime i open the main page i get like 5fps or so cant even use the new slide feature
I think your laptop is letting you know that its 2024 and there are very powerful cheap laptops out there for you to buy. Let it go buddy, just let it go, your old friends needs a rest.
You replied to the wrong person ig
I would love it if there was a working pane where you could keep tracks you are considering using. Often times I'm trying to decide between several variants of extensions. The problem is sometimes the variants stack up and bury the original I was extending from and I have to go fish it out of the library.
What would be more convenient would be a view pane where you could drag tracks over and keep them from getting buried. Or perhaps the create screen could be split so you can browse your library and recently created tracks without leaving the create screen.
Make it so you can get only the instrumental from an AI generation with vocals, or make it a karaoke so you can add your own vocals after
I'd like to add the like, dislike function to allow us to overlay what we like from the song Suno Generated. Like I would like a song, but dislike a portion of the song due to it failing to do something, or did something wrong.
Like you have a good song, but it failed to do a sax solo that you added 7 generations ago.
Hey, Suno team! Are you guys making fun of musicians like me? What's with that STEM function? That's not STEM, that's just forcibly cutting out vocals and instrumentation from a completed song. It's no different from the song splitters and vocal extraction tools that are already out there. Of course the sound quality is poor, the vocals are left with the instrumentation at a low volume, and the instrumentation is left with the afterimage of the vocals. It all makes sense now. You guys have no musical knowledge. It's like taking the hamburger and bun from a finished hamburger and serving it to the customer. It's a completely different story than cooking and serving the hamburger from scratch. Maybe that can fool the uninformed, but the wav export function works the same way, doesn't it? It is not a function to convert a song that was created in wav format from the beginning, but only to convert a compressed low quality mp3 to wav format. In other words, the sound quality is not improved by exporting in wav format, but rather the sound quality is degraded at the time of conversion.
I agree in what you say but I would say it more polite š
Favorites tag and tab is something i really would like, creating a playlist for it is "ok" but more so, sometimes you forget such an easy thing and having a 1 click save to favorites is gold.
can't agree more
I would love to have a video that was designed for youtube dimensions instead of phone dimension. There would be more room for the words and they could be larger.
Now when we post to youtube lots of the video space is wasted: https://www.youtube.com/watch?v=uQjFQTDx02E
Song about the Fermi Pardox and getting life into Space.
Chat is this real? Are you guys just wasting our bandwith with the wav?
gain might be different for every instrument.
Your vocal removal tool is trash. There's literally better vocal and instrumental separators for free online.
Along with the option of making the video in landscape mode for Youtube, it would also be nice if it could generate a new image for each section of the song. When we used to do 30 seconds at a time we had lots of different images in the video but now it just does 1 for the whole song.
The video would be better an image for each section.
Suggestion: SEEDS feature (Like image generators)
like every song has an seed and have freedom to use it again to switch or improve the generated songs, only owners of the song or maybe an option to let the seed public.
Useful when AI generates a great song.
The issue as I understand it is that the way Suno works is not like the way image generators work. As such all a seed would do is give you back exactly what you just created more or less. Not what we're all looking for at all!
So "seeds" as we think of them in Midjourney and other image generators is less useful here.
Suno developers are working on something to get us the ability to get songs that are like previous ones. It's on the roadmap. I don't know how far away it is but they are working on it in the way that the Suno music model works, so hopefully we'll be seeing this sooner than later.
I've been waiting for something like this myself for a long while now.
i think we need a search bar in the trash thing too
Suno should add a feature where you can use the same vocal's from other songs for a diffrent song with diffrent lyrics so you could make albums where they are not just diffrent singers.
You can download both mp3 and wav and check them with a plugin or software that can check the frequencies, I'm sure there's no difference.
I agree with you, but they won't understand if you don't say it that way, because they are implementing features that make no sense at all, such as the conversion from mp3 to wav, or the new STEM feature, both of which are completely useless.
upvote this š https://discord.com/channels/1069381916492562582/1252789683176345641
It would be cool if there was a voice clone where you could set your voice to your account with an option to use that for vocals
It would be cool if we could upload our suno music like an uploader to our spotifys youtube or other music providers
I think editing/inpainting and reusable vocals indeed make the tool most complete in terms of being able to make songs that are fine tuned and more consistent for a creator
i apologize if this has been suggested already but are you guys going to add voice cloning?
Comments under songs! Or some hability to interact with creators/followers
and another suggestion please add a option to empty your trash bin without having to through a bunch of pages.
I would like the option to trash all songs marked with thumbs down.
over-painting would nice. rather than in-paint say replacing instruments in a song
Look at Udio for reference regarding stems. I would like all instruments separated.
not sure if this is the right place for this but - i would like to select two songs from my library, flag one as the start, the other as next, and have suno write a bridge between them
cloudflare having issues
I'd like something very similar... sometimes a whole song is perfect but just middle part is messed up. would be nice to keep the start and end and just redo the middle somehow.
It's tough to get good results from the song radio with the current implementation, all I'm getting is songs that aren't even close song's genre/tags I started from. I think it could be a lot better with filtering for language and genres/tags.
just had a thought. if suno cant generate seperate stems at the same time to seperave out the vocals, what about pan? have the instrumental on left chanel and vocals on the right. would that give a cleaner vocal sound. or maybe switch to a 4 channel wav format 2 for instrumental and 2 for vocals
They should add a "Random" button here - Let Suno generate styles, and such
a song renamer, so that if you make a custom song with a generic name you can have the AI rename it to something appropriate after your done extending changing things and getting the whole song.
Let us specify "longplay" on v3.5 so that when we do "lo-fi" it makes a 4:00 clip instead of ending short or whenever it feels like, tired of doing lo-fi clips that are less than a minute long.
"longplay" in the style tag does kind-of work but I'd rather not waste style tag space for it.
Hey all, it would be super cool if we could have sub libraries, or a favorites folder or something like that? I mostly use Suno to extend my own songs and get ideas so my library is full of tracks with the same name (because im lazy), would be awesome if there was a way to just flag a track as a 'keeper' and have it moved to a separate playlist. āļø
Also - a 'resample' function would be amazing. Instead of extending, downloading, uploading the extended audio. If there was just an option to 'extend from extended audio' āļø
I would love it if it would clone my voice for the vocals.
Can we get an infinite scroll option up in here?
The ability to edit lyrics in a created song and have it generate the music still.
I think we need the ability to modify the lyrics of an already created song or partially modify it without breaking the melody.
I think Suno should add a feature where we can input our own original songs and it uses ai to generate an instrumental to go along with it. For example I have this one song and recorded it as an acapella but it has no instrumental to go along with it, so I would just input the acapella and provides the instrumental with and without the vocals
(Sorry I wasnāt sure which suggestion channel to put this in)
I understand the intention, but separating vocals from the instrumental is something that several softwares have already allowed us to do. This doesn't produce stems and won't be helpful for mixing. The lack of this feature is still the same, we can't use this for mixing. It's disappointing... š
i don't remember where i saw it but i'd really enjoy if we could earn credits if we interact with the community: either by liking other's songs or just listening a certain amount of songs/listening time
I would like a way to continuously generate a song live. Could charge 5 credits per 5 minute blocks or something like that. I would like to use this in my D&D games to generate live music that adapts on the fly without have to swap to a new song
This system is on Civitai. If you like someone else's images, you get credits you can use on creating your images.
1 credit for every minute is a good idea. I like it. And no credits for extensions under, say, 20 seconds.
That sounds good. Doing it in 1 minute intervals would be better and allow it to be more flexible
And it would be good to have an option to complete an extension, making it end, or to have it continuing.
... what are you talking about? i don't care about images, i'm blind. lol
I'm just telling that they have a similar system you mentioned.
I don't know if it's been talked about at all, But the ability to set time lengths would be pretty awesome. So I could specify if I want to render 1,2,3 or 4 mins, this is I found in v3.0 there's some magic in making songs that extends from small parts instead of a full 3-4mins from the get go in v3.5.
how about a crop from feature so wn can end a song without spending creddits only to have it keep generating sounds
Could we get a "Download All" in the Library? Maybe an option for All or just Public?
Downloading one by one, if you haven't been downloading already, can take a long time...
And please, please the ability to select multiple songs in mobile browser for deletion. Otherwise all you can do is delete one at a time
And a sort function in your library, to sort by alphabetical, date of creation, runtime, most plays, most likes
Web App Compatibility
Please consider adding a Manifest.json file to allow the mobile browsers to launch the website in a silo'ed instance when saved from the homescreen.
"short_name": "Suno",
"name": "Suno",
"start_url": ".",
"display": "fullscreen",
"theme_color": "#000",
"background_color": "#000"
}```
You've already done the hard work making it mobile friend so this would be the next step. Your site navigate is solid so lost of the url bar and interface won't effect any thing. Plus we won't get so many open tabs, less chance of navigating backwards and losing a textbox, and it's just the website so no App or Play store to deal with. We get instant updates with webapps.
My suggestion is a simple one, instead of racing to get a 4 minute song made in 20 seconds, how about Suno takes twice as long and works on those annoying whining noises which creep in and utterly ruin a beautiful piece of music. Take your time Suno, no rush!
I think the same! We could complement the audio writing the lyrics and chords in some way to help the ai. Also it could be iterative, if the first approach isn't enough we could suggest what parts ai should modify. It would be excellent. I also have original songs, and I won't feel they are mine unless I could generate also the original melody.
Is there a way to give feedback to the ai when it generates a song with some errors? Like when it gives a wrong sentence in spanish we could correct it. Could this be a good feature for giving precise feedback?
I would love to have FOLDERS to organize my projects.
Why? The longer you are customer, the more songs you create. And the more cluttered is getting the library.
And I'd like to have a "inpaint" feature like in udio.
Why? Sometimes a song is absolutely great, but just a word or a phrase needs to be fixed.
I'd love to have the ability to manually time the lyrics in the regenerated videos ā¤ļø š
When uploading an audio, we should be able to enter lyrics directly, before saving it.
I know it can be done afterwards, but if we could do it directly, the lyrics would be in the video.
Why does it not know it's not over? Why does it just cut off with a minute to spare? Hope v4.0 will fix the issues.
It's a common issue in 3.5.
You can do this tho #š¬āgeneral-chat message
For the Playlist page, please, add a button to change the songs list appearance: in boxes like on index page, or current view as list.
Hi, how are you? I don't know if you will be able to help me here, but I really like this site. I loved using this site. Even my grandmother liked the songs created by artificial intelligence. I'm blind and in the last few days I've been having a problem. I can't reproduce the songs. I can create, but I can't reproduce. And I also came to leave an idea. There is a site that when you like the songs of a creator, you earn some credits. It would be nice if you implemented this idea.
I have a suggestion. If it's possible to add the ability to reuse the same voice model to more songs. Then i would highly appreciate such a feature to be added eventually š
Go vote on #1252789683176345641 by adding your š
Oh okay, thanks for the headsup! š
I was just thinking, is there anyway to show all playlists of someone on their profile list. Right now it only shows 16 and you can only pick new or top to sort them, would be nice if you added an see all to that sorting option
I would love two things to complete the SUNO experience for me. 1. For the STEMS, start separating vocals and music into tracks so they can be more easily split out into STEMS. As it is now, the STEMS sound warped and the vocals still bleed into the instrumentals. 2. It would be amazing if we could transcribe the STEMS into sheet music. Or at least into a Music XML format so we could import into sheetmusic software. Have the ability to do that for the instrumentals and for the vocals separately would really open up our abilities to enjoy our music in other formats. Thanks.
I have a few suggestions about the new theme colors:
⢠Previous theme with strong yellow displayed on the extended songs and the songs tags (version, full Song, parts) was better because it would make it prominent against the black background.
⢠The music composition input boxes (lyric, style of music and title) colors has not enough contrast with the background
⢠The buttons color (including the switches) has the same issue because it is the same color as the input boxes. Would be better to have a dedicated color for these controls.
⢠The overall theme is just black, grey/brown and white wich as far as Iām concerned is not very appealing.
⢠Fix the single clicks bug that makes it select multiple songs.
⢠Iād also add my personal pref to the original Suno logo.
Yeah for perfect separation of instruments and vocals it has to be done from the beginning. Trying to extract vocals afterwards will never work perfectly.
I don't think that separation is possible with Suno; as it was generated together and not separated then got mixed; unless dev could come up with a genius way in the future models
A way to go back to entered text before generated lyrics
Perhaps to dedicate two processes, one for instrumental and one for voice, but have one dependent on the latter. It is a deep technical challenge, but it would improve the sound quality as well as open the possibility of other functionalities, like keeping a instrumental but applying different vocals to it, like the opposite of what we do now with the lyrics being kept but changing the instrumental accompaniment. Just my 2 cents, but just seeing what the developers have done so far, I feel like they can do pretty much anything they put their minds to.
Hey! It would be amazing if the website allowed a little more organization of our profiles. For instance if we could separate songs into categories or playlists, that way if you make multiple genres you could separate them into that category for listeners to dive into different aspects of your work. Thank you for reading ā¤ļø
what tag to use if we want a song to be fast rap
what about a way to seperate the lyrics, and the song in to seperate files ?
would be great if you could see in which playlist a song is when opening the song (but also in library view)
Would be good to know when the video regeneration is done.
Also be able to change the background color or perhaps add an background image and change fonts for the lyrics.
Removal of instructions from the lyrics and also the music style would be good
Instructions to change bpm inside a song. Like [Increate BPM by 10] [1 Octave up] would be nice
There's should be a way to update lyrics for an extended song, so that the lyrics from the separate extensions don't overlap each other. There's been times where i've wanted to extend a song at a drop or a certain part of the lyrics and it gets very cluttery when it just adds the whole lyrics from each time you extend the song
We do have this ability, it's just not intuitively placed. Load up the track's page and scroll waaaaay down to the bottom of the lyrics. You should see:
Ohh! Thank you so much, i didn't know that even existed š
i don't know if i rather should put this in layout but anyways...
i honestly would like to have a better/more optimized player so i could freely move forward and backward in a track without having to always wait until moment comes to check where i should make an extention, for example. otherwise i always have to dl my track just for that purpose and it's a bit annoying
and before you ask, i'm blind, using a screen reader and navigating with keyboard only
Hello, can someone help me
I want to know how can I stretch a single word in Suno AI
Thank You!
Have you used this waaaaaaay?
I'm just here to thank Suno and his wonderful team for finally including the instrumentals in the generated songs to separate the voices from the instruments, thank you
Although I can't test it, I'm already poor and I don't have the premium....
yes i have used it
but its like for 1 second
Is their any other way for stretching for 3-4 seconds
It has been said already, at least by me: it is very difficult to see which song is playing in the list. Please CHANGE the background or the font color of the active song playing (like in Spotify)! It is a programming job of half an hour and it saves us hours daily.
An idea so that users can be notified if a song will be filtered out of Showcase/Trending due to Suno's filters:
After clicking generate,
Maybe scan the lyrics first,
then cross-reference whatever filters Suno has in place,
and if any get flagged, just a popup that says
"Some lyrics or prompts found in your song may prevent you from appearing in Showcase or Trending, would you like to continue?"
And maybe a small label next to songs in your library to indicate which songs are filtered or not
most important suggestion: please hire finally a front end programmer... The changes made in 3-4 months in the user interface could have been done all in 5 days...
Why don't you do it then
any programmer can do it...basic things...
Oh wow look at that, you can apply
I know this 3 months already, old news. But I am not interested in a job as a programmer š I am only stressing out the abnormalities for a company like this.
ouch... I posted some suggestions, but I do not have them anymore.
If an admin can recover my comment then please do. I used a bigger font for titles and it got deleted
Woody is not pleased
I wish we could set a minimum and maximum speed as often it goes too fast or too slow. Also, if there was a flag for "never do anything at all like rap" I would like that.
can we get a bottom margin or padding at the bottom of a playlist, as once you play a song, the music player will cover the bottom track, which if it's a song you just extended and got the whole song on, means I gotta refresh to clear the music player just to put the original in the traysh, not a big deal, just a lil annoying to have to do š
Any idea of when version 4 will be released?
I use Apple macOS Voiceover, in Create mode the track durations are not read out because they display as --:--:-- until generation is finished but the Voiceover functionality in macOS doesn't know this so essentially I can't "see" track durations. Can this be looked at? I am not discounting the fact that it might be me doing something wrong but it used to work with the old UI.
Not sure where to put my point since it is related to DISCORD and the latest changes how to post in the #š£āsuno-showcase channel. We are no allowed to post there only once per hour.
I understand that we should give people more room there and avoid spamming of songs which then move too quickly. But not being able to comment and chat there about songs feels bad. We have built up quite an active communication amomg users, giving feedback about songs, tagging each other, promoting other people's songs etc. Now we cannot do any of that any longer. It feels sad and I would encourage you to get the function back (maybe slow feature also does the trick) or at least provide a room for the users to continue their song-related exchange. I think this is also a great opportunity to create a music-loving creator community. The general chat is nice, but not necessarily about songs, rather suno itself, technics etc.
So tldr: Please give us a room (back) where we can chat about the songs we create and want to share with others.
I have some Python code that calls ollama to run a LLM and generate huge numbers of potential verses. Then I have code that uses the Chess algorithm to rate all the different verses by use the LLM to compare 2 at a time. Then the best of these are often good. If anyone is interested I can put this up on github. Many of the verses in this came from this: https://www.youtube.com/watch?v=VNKY7cwLnFw
Song about the Fermi Pardox and the importance of getting life into Space.
Related ⦠how do some things get listed as āNewā while most do not. What is that criteria?
i kind of like showcase being slower so i can catch up with new stuff people are releasing. A special room for discussing stuff in showcase would be great though.
Please remove the auto-scrolling after generation is complete, its absolutely the worst thing and looking at the message history here, many people agree, anyone who does a lot of generating knows the issue and hates it, it has no purpose but to be annoying, if i wanted to scroll to the bottom i would do it my self, why would it just do it over and over, imagine you go on a website and your trying to read or do something and it just keeps scrolling to the bottom over and over every time you scroll back up
No idea if suggested before. Or if it is a feature and I am blind
But being able to cut away from the end. (Yes I can do that in edit when I download the video, but I can't cut the end of the one on the browser which is the one I make public.
This is an issue due to: Sometimes the AI keeps a song going. It ends it, continues it, ends it, continues it, etc etc till it runs out of 4 minutes, and sometimes I love the final propduct, but I want the last say 20 seconds cut of before I publish it.
when your writing lyrics or anything similer
give a warning when you click a button that would send you to another part of the site
me and many others have lost countless songs worth of lyrics becous of a single missclick
and its a motivation killer and i dont always remember the lyrics i write
Yes. You should basically start from the very beginning of a verse or a chorus.
[Premier Membership paying user] can we remove the option to outright steal other user's songs? Extending from another user's song gives you full access to post the song as your own. why is this even a thing? Am I missing something here (some way to prevent this)? Ever since I learned about this loophole I've been afraid to make anything public on the suno website
i rejoined the server just to complain about the "ideas" in the create tab. get rid of it immediately please. thanks
Allow some way to make it so when we extend a song we can choose if it tries to make it sound similar or not
would be great if the meta tags in the lyrics part would not be shown on the video (why would anyone want to see that?) Now I need to remove and regenerate video, which is a pain and waste of timeš°
Hi guys. Thanks for the fantasic piece of web software. I noticed your model is very poor at creating fantasy tunes (no lyrics) for rpg games I would need. Smth like tunes in Heroes of Might and Magic 3 or something along the lines. Tried your model many times and its great...but not when it comes to fantasy/rpg type of tunes or epic orchestrals like.
I agree with you it has mucked up my screen reader and now I can't navigate the create page, thanks for nothing Suno
Add a new function to see whether the song is included in any public Playlist. It will be very useful to find all open Playlists containing this song and, as expected, all other similar songs.
Make it possible to shorten a 4 min song to 1minute and stil get the same song.
4 months gone and we still cannot mark the songs except "thumbs up". Song management is almost ZERO. Even Bill Gates programmers were better and faster...
an option to extend using the existing prompt/seed so when you have ai generated lyrics the extending should work off those and not completely switch or rely on manual input.
Don't know if someone asked for that already but ... Would it be possible to get an option to just empty the Trash?
Add something to make a tag more or less important. Idk with numbers. In my opinion it is gonna greatly increase the control of a song
I cannot understand why they have mostly bad ideas...most changes are annoying and not helpful.
they have poor audio quality yet decide introducing an unwanted change to the UI is worthy of time. Suno have their priorities wrong and as a result of this change have made the site less accesible for the vision impaired.
I am also not a fan of the ideas tab in the main screen,I think it could be put elsewhere like the explore tab has genres. Plus the auto-generate lyrics is essentially the same thing anyways.
Iām also wondering if it would be good to have the creators listens of their own songs not be tallied on the listen counter so we know when others listen to our songs.
Very low quality. In the new feature I believe the AI āāneeds to understand the structure of the instrumental and then rebuild it with better quality.
I really liked the new feature. Now it's just a matter of time to improve it.
"Library" could use many more filters beyond "liked" and "public", but the one I would put on top of the list is "Clip 1 Only". When I'm flipping pages to find something I can't get with search, many pages are just extensions from the same song I'm looking for, and I could skim much faster seeing only Clip 1 (and then drill down into an extension, if necessary)
why do we still have this strange 9:16-ish album art and the only option to upload is square? this is making the selection of decent album art slightly more difficult which is extremely upsetting to me. either we need the option to select the proper ratio when uploading, or we need to go back to square album art. make it make sense!
There should be a feature in the "Style of Music" box that allows us to save combinations of words under custom names. For example, you could create a retro style with "1950, ragtime, avant garde" and save it as "Retro 1950s". Would be a neat way to save/share crafted music styles.
Moving something to the trash should not automatically make the page scroll down to the bottom (in the create tab.) I scroll up in my list of songs, delete, and then it shoves the page back to the bottom so I have to scroll back up to delete. I simply don't see the utility in the automatic scroll.
we definitely need a instrument conversion where you can use something as simple as a synth to have the ai recreate the pattern with like a guitar
Agree that it shouldn't move the page, but just so you know, there is an easier way to delete stuff. You can select multiple by either clicking one and shift-clicking another to select everything between, or you can control-click multiple to select them individually, then when you select to delete them, it will do all the selected ones. It functions like selecting multiple files in Windows Explorer (I've never used Apple stuff, so I don't know if there is a comparable function there).
i miss being able to toggle repeat and autoplay on my phone when i am accessing the music i have made. why was that taken away for the sartphone?
Oh my god you're saving my life with this š
Thank you ā¤ļø
one of the best features for song management would be to add folders (not playlist!) in the left column (like in the spotify) and to be able to drag and drop the song in one of the folders with the mouse. A single huge song list ("library") as it is now, it is a nightmare. I wonder who from suno had this very poor idea, specially for the year 2024 and with so many good examples out there in the world.
Add an ability to download songs you multiselect so when batch making songs it is easier to download them all at once for videos
Also it would be nice to have the stems like how it is in the daw instead of separating from the mixed version
The reason why it's this why is because Suno doesn't have stems
It doesn't use them, at all.
It's the master and that's it.
thumbs down should refund credits, even if it's half back. Losing credits on bad songs feels like crap.
It would be neat if remixing and forking other people's song was kind of encouraged with Suno features (probably after better permissions?) For example:
I made this: https://suno.com/song/9f35c0ae-bd30-4d07-a5a8-2622376aa0b9
The "Durt Mudda" made a Spinoff (with permission): https://suno.com/song/c00faf15-2348-4b02-a4e1-dad164275a3a
On the song page, there could be a publicly visible list of all the spinoff songs that other users made from Extending it. This would be a fun reason to allow and encourage people to collaborate and riff directly on each other's music.
It would be really good to be able to regenerate parts of a song. Both using minutes and seconds and also if possible regenerate i.e. Verse 1
how many times do i need to regenerate a video for it to work. you are wasting my internet data. https://i.imgur.com/JydUHGM.png
This would be fine if remixing and downloading of someones songs was opt in. No one should be able to go to anyones page, extend their song and say its theirs. I shouldnt be able to go to your song, press Ctrl+U to get download link for song, and download it without you knowing or having to give me permission to download it. Protecting users and their creation should be number one priority. Even if your songs are private, they can be found on google by simply searching for it. Its going to be really hard to convince people to stay when they start finding out someone's been stealing all their songs. On top of not having any viable proof of ownership or creation of said songs.
Seriously you need to get your act together. As an example your AI got no clue what 80s New Wave, Synth-pop, Post-punk, New Romantic etc. is. Every try comes out sounding like a modern pop song (and yes I do know how to write good prompts and as a muscian I also know musical terms etc.). Oh and don't get me started on the vocals. Did your AI only listen to one male and one female singer? Almost every take ends up sounding extremely similar with the same articulations, phrasing and vocal style etc,. You are losing the battle to UDIO if you don't get your act together and that is a real shame as your sound quality si much better. But the variation in the music and vocals is a huge letdown.
you'll probably want to use v3 currently for generating styles like that as v3.5 stylistically is much more heavily influenced by mainstream and pop music currently. all of the genres you listed are things i've been able to achieve in the past consistently by listing genres and the moods i need, but 3.5 is probably steering it the wrong way if thats what youre using atm.
This will happen. Maybe not with Suno. They might not get on the train. But it will happen with other AI. Just keep your eyes open.
Maybe try duplicating your V3 prompt in V3.5. Like copy and paste it so it's 2x or 3x in the V3.5 style prompt. (Skip mainstream genres maybe.) Sometimes just that can kick a sub genre in to gear.
When you extend a song, you cannot see down there which model version it was made with... so add the version in that line with the extension informations
There are some songs where the intro has a small part of the voice like "mmm", "ahhh". If you extend the song it will take the same voice which is not so good, because you get restricted to that voice. But that voice could be from the background singers, not from the lead singer. So it would be nice to have an option when extending a song to take any other voice by random, so you dont get stuck to that one voice.
Affiliate / Referral program would be nice.
There's a couple different ways I've seen stuff like this handled by different websites / services, etc. Which implementation do you think would work the best for Suno?
The first thing that pops into my head would be if someone signs up using your affiliate sign-up link, or if they clicked a "sign-up" button while listening to one of your songs, if they opt for a paid membership within a certain amount of time, you would receive a certain number of "Extra Credits" that don't expire.
Just for brainstorming's sake, do you have any different ways you would like to see it handled?
Are there any plans to implement a feature to share songs without allowing others to āreuse promptā or āextendā within Suno? I feel like a lot of folks create stuff and then share out through YouTube and Spotify, but that traffic could be kept and directed on site if this feature was implemented. Sorry if this is redundant and has already been asked/answered!
I can already see the day in the future when SUNO's AI will reach the level of being able to change the lyrics of existing songs without changing anything in the pronunciation or rhythm of the words and without changing the melody.
One Direction ā What Makes You Beautiful (Official Video)
Follow on Spotify - https://1D.lnk.to/Spotify
Listen on Apple Music - https://1D.lnk.to/AppleMusic
Listen on Amazon Music - https://1D.lnk.to/AmazonMusic
Listen on Deezer - https://1D.lnk.to/Deezer
Listen on YouTube Music - https://smarturl.it/OneDirection_YTMusic
Check out One Directio...
One Direction Parody!
See the original- http://www.youtube.com/watch?v=QJO3ROT-A4E
See Behind the Scenes- http://youtu.be/wNanusH6L8o
Check out our second Channel! http://bit.ly/xtPqZI
Subscribe to Awesome! http://bit.ly/y2zUpn
Written By Mark Douglas, Todd Womack, Bryan Olsen
Directed by Tom Small
Boy Band Members:
Jason Michael Miller
...
Bumping for visibility
it be amazing to be able to rate own creation and others with stars š
Came here to suggest the same. Extending or reusing a prompt from someone else's song is something few people are interested in doing (unless they are setting out to work on the song with the person per some sort of agreement) and fundamentally misunderstands how many if not most users seem to feel about their lyrics. Allowing it should not be a requirement to publish on Suno, and a substantial number of people who have made their songs public on Suno would probably not have done so if they had realized that feature was there.
Please add a Uploaded Category tab to the library cant find some uploaded songs i want to extend..
who tells the UI programmer this another basic stuff: ALL sites in the world who manage a list with many entries like the library in Suno, they don't only give you the possibility to jump to a certain page from the start (and not after user begg for it for 3 months), but they ALSO show the user HOW MANY PAGES are in TOTAL ! This function is missing. The lazy UI programmer and his manager should be finally changed for someone competent. This is kindergarden.
You can type the page number you want to go to in between the two arrows
I know, but this came after 3 months... and the total number pages is not displayed yet.
Oh ya also you can just keep pressing the right arrow and go to pages where there i no content
Sould fix that too
They all know about this bug, but the lazy UI programmer and his manager dont care..
they do a single change every 2-3 months...so in 20 years they might remove all the known bugs and implement the basic feeature any non-Ai site has for decades...
I mean i can live with the bugs but i hope they mange to create voice models that dosnt sound like a roboter sometimes in the near future
the problem is only that most bugs can be fixed in 1-3 hours... š this is the scary thing.
Also implement all sub genres i try to create a glitch hop track but it creates something completely different same with slap house or any other sub genre
As long as they put their programming skills into making the ai better im happy with their service but i know those bugs shouldn exist
you dont understand: a cook is not the waiter and viceversa. The devs which programm the UI cannot be the same as the ones for the sound or for the AI model. There are different things they have nothing in common and can be done separately. Even for Android or Ios there are different devs
I dont know how their team is setup but sure if they have a team only for the ui they should maybe overthink if they got a good team there
no, for UI is nobody. That UI "team" is only one guy at the most, not a team, and he had to work for something else (android ?). This is why we dont see nothing !
Adding new things before fixing those minor bugs should not happen
Idk who need the "Need Ideas?" feature, i mean those ideas are in everyones head when they want to create a track
Thanks for the option for stems, is there anyway to get the unmastered versions?
Some feedback for Browser
Really some of this could go into application as well. Bringing my day job home with me, and looking at Suno the same way I do apps I do for work. š
Lack of Analytics
Suno is providing a platform for song creation and distribution, but I have no meaningful way to track my growth rate. This is important if I want to treat Suno as serious platform. I'd like to be able to answer basic question like:
- Fan acquisition:
a. What tracks lead to them following?
b. Where are they located at a country level?
c. What's my fan acquisition over time? - Play time per song:
a. What's the drop off rate?
b. What's the % of fans vs non-fans listening to a song - What genres are performing well
a. In what countries are they performing well - Are my tracks being shared?
a. From what countries are they being shared?
b. To what countries are listening? - Profile views
a. What songs are leading to profile clicks?
b. How many profile clicks am I getting?
c. Which countries are viewing my profile? - Playlist likes:
a. What tracks lead to them liking a playlist?
b. Where are they located at a country level?
c. What's my playlist likes over time?
Profile Page
- I have no way to know what clicking the play button is going to do without clicking it. It appears to play the most popular track even if I've filtered by newest track
- If click to favorite a track I see it favorites my track. If I click to favorite my playlist I'm taken into the playlist. This is inconsistent.
- Sharing my profile normalizes the username which looks pretty weird. For example my profile is https://suno.com/@xxxxxx but when I click share it creates a link https://suno.com/@XXXXXXX both work (hopefully not a real profile), but if a user cares about branding, this is negating a user's chosen username format.
- Why can't I click on my fan total to see my fans, and why is it called "fans" on profile, but "followers" in my library view. It's inconsistent, and I prefer "fans," because followers are in a cult.
Left Navigation
I personally hate "Explore" being on the nav for a paid account. It rips me completely out the experience, and I find myself frequently clicking it by accident when I meant to click "Library" or "Search." It feels like something you'd want on a nav for users that aren't signed in so they can see what Suno can do, or maybe for free accounts.
Home Page
The navigation to expand the list of songs is really confusing, and I have no idea what they do without clicking on the options:
Under Following the expand is called "Go to Feed," so why not call this section "Your Feed" instead of "Following."
Under Summer of Suno, Vol 1 it's "Go to Playlist" but I have no idea what I'm seeing is already a playlist because it looks like every other list on the page.
The order and naming conventions of sections feels so haphazard. For example, different musical genres are labeled "Categories." "What category of music do you like to listen to," asked no human being ever. It looks like the iOS already realized this, and it's calling the list of genres "Genres," and not categories.
Create
The need ideas section is over whelming the page. I need ideas never. I want the ability to collapse and hide this section by default for my profile. I only ever want to be in custom mode. I feel like I'm working at mission control when I open this page.
On the help button for Lyrics you're using the Oxford comma, but on Title you're switching using AP style. Generally I think you'd want to pick one style or the other for consistency.
Library -> Followers and Following
This page isn't responsive. Also it's called "Followers," but in the profile the same concept is "Fans."
Playlists
I can't understand what the sort order is here. It's not likes, it's not alphabetical, and it's not chronological.
Search: Songs
I don't understand why in some places in the UI you're collapsing the numbers of be readable (e.g. 300k), but on this page it's 300000 (which is far less easy to read)
Example of comma inconsistency with lists. Is it minor? Sure.
I was looking forward to it because I had a get stem. Contrary to my expectation that it would be completely separated from the instrumental I had hoped for for a long time, it was very disappointing.
It is worse than separation through UVR.
Does the SUNO system create both inst and vocal when creating a song?
Can't we separate the two.
I'll wait a little longer. Suno will update it..[trans kor->eng]
Not really a browser issue, but I do wonder if Suno wants to use what appears to be sequentially assigned user ids instead of UUIDs for accounts. That leaks meta information about Suno's user base size, and also high value accounts based on a lower assigned id. Given this information doesn't seem to be used anywhere, this might be just leaky information the server shouldn't be returning at all.
Suno does not have separate generations for anything that makes up a song. The reason why they can't give the stems is because they don't exist
It's master only, which is why it has to be broken apart afterwards
In the end, you have to take the finished song and then break it back and remake it.It's hard to make it completely your own.hmm..
Include an option to manually change the RANDOM SEED in custom mode. This way I can make changes (correct lyrics) without changing the whole sound.
So its 11 pm and all my credits expired? I've still got an hour left before they expire.
No where on the credits page does it say what time zone they are in so why wouldn't be my own?
Please fix the random numbers bug when extending songs (even before the 4 minute mark) and the warbling effect that happens to voices half way through songs š
I lost the adjustments I made to my lyrics again while in create mode. I finished my changes and instead of hitting create, I popped out to rate the previous version. When I went back to create all the information had been reset to the default state. I lost the 9 minutes of tweaks, changes, adjustments and micro experiments I made.
My intellectual take:
I know this is my own fault. I know better. Itās happened before when Iāve moved out of create mode. I usually copy and paste to a note file, but I simply forgot.
My emotional take:
Iām pissed.
I shouldnāt have to do a copy and paste backup every time I leave the create tab. The information we type in under the create tab. Should stay there until we decide that it shouldnāt. There shouldnāt be anything that causes what we type in under the create tab to be reset to the default. I shouldnāt have to remember that this can happen. I should just be able to know that what I type in will be there.
Iām pissed.š
i totaly agree with this cause it happened to me too. i mean quitting the create tab by accident and then coming back to it to find it had reset. hopefully most of the time i just come back to the previous page in the browser so that seems to come back where you left
Another set of things I wish I could do:
- As a user with a presence on many social networking sites, I want the ability to share these links from my profile, so that people can find me on other platforms.
- As a user concerned about security, I want the ability to protect my account with 2FA, so that I can add an increased layer of security to my account.
- As a user that cares about a standardized online presence, I want the ability to add a header image to my profile, so that my profile can look that same across the Internet.
- As a user that wants to collaborate with others, I want the ability to display a link to the song I've remixed from the remixed song, so that I can support the original creator.
People at Suno.com,
please implement a routine to batch download all the songs on my profile.
It could be a block of all the songs generated on a specific date or a specific interval of days.
In the 21st century, having to download music one by one is the end of the line.
As we all know, the style of music is set when you made the music, however, when you extend the existing music, the original music's style tags is fixed, the only way you can determine if the style is change is through the video, so, could it be possible if users can modify the style of music when the full song's made so that confusion can mitigated?
only on song details page.
they should add an option to add custom adlibs & harmonies, & more high notes
like a prompt code with longer characters
an instrumental arrangement feature would be amazing. Once a song you like it made or you upload audio you can do something like ā0:10 to 0:30 guitar onlyā or ā1:30 to 1:40 piano and violin onlyā. That way the user can have more input on how their generated songs are composed
yeah and also the ability to make live arrangements with drums & that, and also extra vocals for the live
I told them this long time ago too, nothing happened.. AI age and manually download doesnt match
we need something other than a slightly different grey button to let us know that suno detects "vocals" in our audio uploads
my fault for sure, but i've wasted a few hours of my life on things that I didn't realize I couldn't make public until it was too late!
a button near the play button to increase the play speed to 1,5x or 2x would be great for saving a lot of time when listening to song for different purposes... whatsapp has this š
It totally depends if you want actual referral program or affiliate program that pays $$$. There are lot of good existing affiliate dashes that allow ppl to generate their own tracking links and coupons. Personally i prefer cash option, but most or some AI tools have both aff and ref programs in
ok.
So let's start a petition to collect thousands of signatures asking for the implementation of this feature.
When a generation fails, instead of refunding the credits for that single song, just give an option to regenerate the failed generation
Suno, when are you going to release the feature that allows users to input their own voice to sing the lyrics of the songs they create? This would be amazing and would attract even more people to use Suno AI! I believe everyone agrees with this, and it would be great if everyone requested this feature from Suno AI
I know it's been asked before, but we really need folders. It's impossible to organize anything, and when you've been making songs for awhile, you just have mess that you can't really ever go back to.
Edit and or Extend the beginning of a song, or any segment for that matter.
Please upvote.
Comment section would be nice.
GIF covers could be cool too on mouse over.
An alert as to why some songs will not be pickup in trending (forbidden lyrics, cover img, etc) with the reason why would be nice so you can learn from it
Like button in the discord share link embedding. (same for other platforms x.com etc)
Change the embedding in discord from vertical rectangle to horizontal rectangle, less space will be used
Default remastering modes to clear up audio
Are these songs generated similar to open source diffusion programs used for images? If so a seed number would be very welcome. Being able to recreate a song from scratch in the same seed it was produced with and altering a tiny character that sounds wrong like the words: mate pronounced as MA and TE, Lives, pronounced as living instead of laives. It doesn't always get the word right being able to help it to recreate and not having to extend it would be a very welcome feature. It also gives users a lot more control with extends too to keep the same extend in the same seed or the next extend after 2 minutes to not deviate from the ongoing flow of the song
I like most of your ideas but gif covers would probably cause lags and delays.
Fair, got the same feedback from others. An on/off switch might fix it but users. Probably wouldn't know where to find the button
Yeah that's not worth it. It be a bit to much.
Maybe gif avatars for profiles but for covers isn't a good idea imo
Agreed
yeah im at about 12 regenerations for my current song. no matter what i try it wont take the updated image... it seems there might be some kind of internal cache..... ive had this issue about 3-4 times in the last few days... i take a break try again the next day and it finaly takes the proper image... its kinda annoying and a waist of time.... i hope they get this fixed.
š!
Thought I'd toss a few in here.
A search function once inside someone's link, so they can search the library of that user?
Ability to sort by alphabetically?
Sort by most liked?
Sort by most played?
Inside your own account, ability to search hidden songs?
I've already emailed this, but saw discord after.
The credits are too expensive. 2000cr for 16$ when we get 2500cr for only 10$/month!? I would buy more credits if they were half of the current price to be honest. (I currently go through 2000cr in about 3 days) š¤Ø
I don't think it's that spendy at all tbh
Or depending which subscribtion you already have you make credits as cheap as . The subscriptions factor. Monthly plan 2.5k = 10 euro for 2.500 or a factor of that. Monthly plan 10k = 30 euro, so 2.500 for 7,50 euro
Proplan 2500 for 8 euro
Premium plan 2500 for 6 euro no plan 2500 for 20 euro. Would also be nice to make the euro price cheaper and same value as the usd one, because Europeans always pay more then Americans with all of the services from games to music to TV etc already. as the euro is usually higher in value compared to usd by like 7 percent. 10 us is not 10 euro it's 9.3 euro. Especially with our lower salaries in europe in comparison it makes everything more expensive
It is not unusual that extra credits cost more money on other platforms as well. I guess it is because they want to offer the subscriptions a little more affordable, i.e. the extra credits cross-finance the subscriptions.
I bet a lot of users didn't use all their credits in a month.
I wish there was a delete fail safe for liked songs. How I like to work is:
- Generate a bunch of intros.
- Then the ones that will work, I'll give a thumbs up.
- Then I'll work on a track
- Delete the ones I didn't thumbs up.
The issue is there might be 100 generations, and within the hundred there are 2 thumbs up. If I miss them, and then end up in the trash by mistake, it's nearly impossible to find them again (especially if I don't noticed within the undo window). If instead the UI said, "Hey I noticed your deleting something you like, are you sure?" This would solve the problem. Alternatively just let me filter by likes that are in my trash.
the proposal is as follows:
add to the generated songs the ability to separate vocals and music, thus the quality of the instrumental component of the track will be in good condition (ideally at the stage of downloading the track), please pay attention to this!
its probly already been suggested BUUUUT:
Request: change connection type 'twitter' to "X" and try fix the bug on adding x.com connection....
The Stem separation is after The song is generated. It's the same thing we can do with other software. They need to train the AI on separating the instruments as it's making the song like being tracked in a recording studio(from a yt video)Please we need this
Why is using the lyric 'dead' such a big deal? I like to write horror music, but everyonce in a while a word is censored. Can we just not censor words and enjoy free speech?
Hey boys and girls at Suno, could we get an add on soon where we can ābetter manage and organize our music, for instance, āA wonderful work area for making/generating music, as the main work area, then instead of just trash, we have trash for of course songs that will be deleted, and even a delete permanently button on songs in the workplace so you can get rid of them entirely, the ones that donāt turn out properly, then, have a create a new folder feature so that we can put finished songs away outside out work area, so they donāt just pile up in work are, because there either just piled up in the work area, or in the trash, and it sucks to sort through them in the trash with many other unused or wanted songs, and it wouldnāt hurt to throw in a ādelete all button in trash, so you can just rid all garbage at once, then of course with new folders, the finished songs can be safe in their own folder! š
It's probably in the specific context of the lyrics. Ask ChatGPT about it. ChatGPT is the censor.
It would be nice to use SEED #'s to include vocal styles or specific tracks in other productions. So users can develop VOCAL libraries to use with different styles of music. Also, the library page should have a section for user uploaded files.
i think it could be fair to regain credits when we delete songs/clear our trash, like one credit per clip. cause if they are in the trash it generaly means it is a bad and disliked generation. or if one credit per song is too much, then one credit per pages
Agreed,or maybe they could charge the credits in order for us to "unlock" a song's/generation's features like download/stems/extend/share etc. (locked = listen only?)
mmh i don't know this sounds very restrictive. five credits per generations can go very very fast, why chare more just for some features?
thank you Suno dev for waisting millions of hours of the users by using 2 klicks and 2 menus to download a song. Not to talk about having to download each song manually ! Time is also for us money and energy!
Like some other AI's prompting interfacing website, it would be great to add a button "Use previous prompt"
Ded š ssssssssh
Suno does have that feature, it's just in the track menus now as "Reuse Prompt"
Great ! Thank you mate š
So why you don't prompt them?
I do prompt, but the character limit is very limited, only 120! they should let us type more
re-use seed so lyrics can be edited, being able to change lyrics would also be a plus
also, would love to be able to extend from a specific split dynamically, as lyrics can blend between sections
Being able to Extend Backwards in time to take a great clip and make a different intro for it
Chatgpt can be very puritan in nature. Oh well
how about a group plan for projects where an entire project can join and reap a benefit or reward suno as a radio service

offer suno too web3 projects as a radio server for example a founder says yes i want suno then founder pics their pfp < or nft and that would be the groups pfp < of the package they took soo if music is released under that package it apears with the nft as the pfp
advertising for that web3 project more creds for holders of the nft and that user in suno and ofc profit for suno
soo holders can then make project specific music or host radio on that respective project

the TTS doesn't know the difference between breath and breathe
"Full Song" filter
Keyboard keys for NEXT/PREVIOUS song instead of the mouse would be also nice.
Only space for PLAY/PAUSE is to short-sighted dear forgetful UI programmer.
It does, but some of the "singers" may struggle with the pronunciation. It's probably very dependent upon the styles and the voice the track ends up with. My test worked fine: https://suno.com/song/27b49fe5-a35d-4e3c-8843-92881e73aa69
well, linguistically, "breath of fire" shouldn't say 'breathe of fire"
I agree, tho my response still stands. Suno mispronounces lots of things, but it seems to be very much dependent upon the voice the track ends up with. https://suno.com/song/50d6c8a2-57e4-4fcb-a403-6b6f8c949694
Breath of Fire came out fine in that
Rebumping for visibility
Stop sending songs to suggestions, instead, send them to Suno showcase chat, #š£āsuno-showcase
Relax. I wasn't showcasing those songs. They were a direct response to that user to demonstrate that suno can indeed properly pronounce "breath"
When I first saw those red dots that briefly appeared next to unplayed tracks (for all of a day), I thought they were to indicate that the tracks had hit the filter. š
The possibility to use a negative prompt would be nice to have more control.
No clue if that would be possible, but using one of your own Suno song as source for a new one for its style with other lyrics would be nice.
I would love a CUT HERE button. So many times the song keeps going after the perfect ending.
At least you can cut it with the extend feature.
That doesn't always work. It's a band aid at best. I would love a Cut at timestamp xx:xx
Yeah, that would be better. Generally a song edit feature would be nice, even to recreate a part in the middle of the song.
For example I make mostly songs with a female chorus, rest male, sometimes the song is great but the chorus not female. Or parts that should be male are also female. XD Here a edit feature would help, you can even use it to insert instrumental parts after creating the song.
I wish there was a module for co writing, working in the same environment for a track. Screenwriting apps have this. And (previous) lyrics frame next to the new creation.
where is the style exclude option? is this a joke?
hello. Can I suggest an idea for suno ai here?
Familiarize the AI with music and rhythm terms to accurately follow requests like "6/8" and other stuff
Also here's an idea so failures aren't as frustrating:
Generate just 10 seconds of the song preview then if the user decides to continue make the full song
If the user thinks the song is a failure refund some credits
Would be nice if can edit songs like cut off part of song off
Yeah, that would also reduce the used AI computing time and lower cost. I spend 400 credits on my last song and gave it up because the AI ruined it from the start (for example [Intro] is often zero understandable because too quiet or the beat so fast or slow, that you know the rest is garbage).
Make the voice seed (number) visible so you can reference it and then also add a Voice Seed field (or allow it as a tag) so the user can reuse a specific voice.
you may try to add some meta tags with extra style. Check out strauss posts and his metatags in the videos, I have also posted my extended tags. The thing is you can add the style tags directly into the lyrics. What I sometimes do, I type in the prompt some extra instruction like: Interpret lyrics like ... or Interpret something in brackets like ...
No worries, I get that now! š¤
If I can make a suggestion, is it possible to have a "Remix" function for AI songs based on audio samples we uploaded? Where instead of having the AI extend a given song from where it was left off, it can change the style of what already exists to the user's prompt? I think this can make it easier for people that don't really want the AI to deviate too much from the original song, so they can extend it later when they want to.
I think it would be a great feature for free users, it's even better than a remix. Here's my idea: you upload the original track and a second track from which you take samples. Artificial intelligence then overlays these samples onto the original track, creating something new under the beat. It's just that I've seen it repeatedly in tiktok and if you want something that isn't there, then suno is a great solution! (put positive reactions if you support my idea)
hi yes remix would be great and can you extend the v3 to 4 minutes, its aweful this v3.5 , wont listen to prompt at all
and no matter how many times prompt for female vocals, keep getting male vocals, are these guys even listening
Suggestion: A visible indicator of whether or not I have ever played a particular clip. I'm mainly thinking about this in my own library, but I wouldn't mind it on any song.
I would really like to make a medly of multiple different songs and string them together. Extend can sometimes work, but it has inherent flaws (degrading audio quality the more you extend, the prompt has a hard time changing the genre of the song between one extension and the previous song and requires a lot of compute.)
My suggestion would be for a feature where you can pick two completely different Suno songs, and have the AI string them together in some meaningful way. "Stitching", I suppose this might be called.
Is there a chance points are going to start rolling over to the next month? I often have more than a thousand points left by the time the next month comes. This is the second month that this has happened. I feel like I'm not getting use of all my points. It'd be nice if they would roll over to the next month. Wasn't sure where to post this. Sorry if this isn't the right place.
Please consider refunding credits for extensions up to 15-20 seconds. When I try to end a track, there's literally nothing there worth credits.
if you buy "extra credits", which gives you less than half of the songs, they wont expire
++
an important suggestion: what if Suno would finally start to implement suggestions instead of gathering 10% of them?
Yes, a song trimming feature, please!
are there any tips and tricks in using suno that yall can share to make better songs?
Yeah, it would be great if unused credits rollover to the next month.
Search function in the Trash bin
Suno should guard against a seemingly no limit on page size requests on their API. Generally speaking you don't want the backend to allow for page_size to be infinitely large or infinitely negative because it let's someone extract an infinite amount of data which depending on the architecture could really cause havoc. This size should be validated to ensure this value can be >= 0 and <= some desired max size like say I dunno 250.
Any hopes in the near future for a captions export functions? So we can export the vocals along with their timing? AKA SRT FILE
Sorry to bother. I purchased a Pro Plan to renew last year, but my region no longer supports account login, how do I cancel auto-renewal?
I still deduct $10 per month.
Please send an email to billing@suno.com about this.
One of the hardest things to do in Suno right now is repeating elements. Copying and pasting blocks into the text editor SUCKS. We need some way to indicate when we want to reuse lyrics for verses, chorus ...etc š
Most of the time the ai does Not take your prompt seriously if it has specifics
It will just probably ignore the specifics and go on its own way to make a song that is far from what is requested
either more advanced stims to make looping songs or simply a option to loop the song (ending sounds just like the start or selected time)
The latest update has again broken the ability to download in wav format!!!
I have a suggestion. Would it be possible to get the music sheets from the songs?
That way musicians can more easily learn the songs themselves and adapt them more easily
**Suggestions **
Updating my previous post.
Got some new ideas today.
1) Set playlist order not based on current criteria. (I don't mean the song order in a playlist but the order your playlists are displayed on your profile)
2) Add filters for Suno version used. Include an on/off parameter for songs to declare if they are finalized, allowing filtering of finished or unfinished songs in playlist searches.
3) Include an option to "hide" songs from profile overview. Older songs with high views can still be accessed in a playlist named "Old Songs" but won't appear in the profile overview.
4) Enable song trimming, allowing users to cut the start or end like YouTube's feature.
4.1) Introduce the ability to adjust voice and music volume using key points on two tracks, enabling fine-tuning of voice and music balance.
5) Introduce the option to self-boost songs for credits. Include a section on the Suno homepage for boosted songs, where users can pay credits to have their song highlighted for a set period, with a queue system and estimated posting time. (Some chatter about this point after my last message)
6) Provide a clear list of supported statement tags in the help section. Investigate issues with the [End] tag listed on the Suno wiki.
7) Implement track layers similar to Photoshop, allowing users to add additional vocal or instrumental layers on top of the generated track.
---> I havnt used the stem function yet, but from what I understand we can download the separate tracks and edit them outside of Suno... but we can't regenerate a track in suno with corrections to each stem, or am I wrong? I have a song I would drop the sound of the beat in a few sections so its less distorted and regen the song maybe? (my Scorpion / mortal kombat song if anyone is interested)
8) Enable gifting of credits or monthly subscriptions, as well as credit trading between users. (I would expect to only be able to trade the extra credit bought and not free credit obtained or subscription credits)
9) Allow for switching between male and female voices, possibly using designated tags such as [Voice Signer1], [Voice Signer2], [Male Voice], and [Female Voice], with support for multiple signers. Investigate and address any inconsistencies in voice switching functionality.
10) Somewhere in your profile page, Would be nice to have a history of your songs that hit the front page (Trending) This can be motivating. I'm sure some of us out here may of had a song trending and didn't even notice.
So a history page with:
- Song name.
- The period it was trending for, so we know how long it was trending.
- Top position it reached in the trending
Also to add to this.. when a song is currently Trending, on that songs page you should have a temporary tag "Currently trending"
11) If your on the page of a song that was deleted, you should be able to restore the song from the song page. having to navigate 80 + pages in my trash to find it when its already open is kinda annoying.
12) Implement a search feature to the trash maybe? or make it so the last song deleted is on page one. its very unclear how this trash is ordered.
13) *NEW today* Allow to extend a song without pulling previous parts before, if wanted. So when I do "get full song" that would only pull part 2 and 3 for example and ignore part 1.
Lets say I'm on part 12 [Extreme example] I could say in a drop down of some sort only pull after part 6 and stop at part 12.
14) *NEW today* A messaging system to contact creators.... If I wish to reuse someones unfinished song how am i suppose to ask for permission to use his song if I have no way to message him/her... we can extend other people songs.... but we are not allow to monetize them without getting permission from the original creator no? how does Suno sees this extending someone else song? if its not allowed why allow us to do it? I'm unclear on this.
I'm still waiting for Repeat function on playlist. Definitely a must have.
a lot of times i get a boring intro.... but i get an insanely good chorus... so i want to keep it, and edit the start of the song... but that is impossible.... a "front edit" or "front re-generate" function would be INSANE.
Better organization of the 'Library' in browser. Better use of space, section for user uploaded tracks. Thx
Please add "Artcore" genre to Suno.
What is artcore?
In general, it's an experimental genre of combining orchestral and electronic instrument.
Most common trait I noticed is that orchestral lead tend to be more complex and chaotic support with electric sound
Not sure if I descibed it right, though
You can look up channels like Yo Kaze and Afflictive to hear Artcore musics.
could we do reference like what midjourney does sref?
i mean artwork and songs have the same copyright and midjourney is allowed for some reason
wouldnt suno be allowed too?
or am i missing something bc i swear i heard that art (painting) and art (music) have the same copyright
When the Paintin function comes?
v3 extensions should extend up to 2 mins just like v3.5
This is my recommendation , I feel like for people like me that listen to the entire songs from the results , instead of just looking at the browser , there should be a little interactive music Game you can sort of play while previewing your own tracks & deciding which Variations are the best , I just was noticing myself Doing a (& don't judge me) Tycoon on Roblox and I was able to listen to more of my Tracks in a go without getting too much Ear Fatigue ... So possibly have something you can do with your mouse while Previewing a Lot of Tracks
please add a remix function, as I saw a good potential for this in the extend section
please add more payment methods
can we have shuffle for playlists
The ability to have folders and put different projects in folders would be extremely nice.
There seems to be a bug? Since ~20 minutes the create tab shows me only 1 single song, the last one, and not all anymore. The rest of the list is empty, the songs are still in the library tab.
Edit: maybe I had too many downvoted songs behind this one song, after moving them to the trash the create list works back normal.
Edit2: so it looks like if you have negative voted songs in your create list, they get not fully ignored when you reopen the create tab, they block songs from be listed.
Generally a direct delete option would be nice. Generations are often so bad that you only want to delete them and getting them out of your mind. XD
Or when you want a male and female song and the AI generates only male or female > instant delete. The voice sounds bad, it sings the line horrible, the beginning is very bad... there are enough reasons why you directly want to delete and don't move them to trash and delete them from there. That are no pieces that could be any useful. Main reason why I say that: my trash list is way too full and it is a lot of work to clean it up.
Same here
It's already back again
But only for short, here it is again. XD
Hi everyone,
I've tried uploading a recording of myself humming a song to Suno and used the extend feature, but I didn't get a good result. Is there a better way to turn my humming into a polished song using Suno? Also, is it possible to add instrumentals and rhythms to a recording that only has my voice?
I would like Suno to arrange songs where both the lyrics and the melody (composition) are mine. Does anyone have information about Suno's development plan? I am imagining being able to arrange songs where both the lyrics and the melody are mine and make specific changes to the output (e.g., change the part between 15s and 55s like this...).
Any tips or suggestions would be greatly appreciated. Thanks!
Best regards,
The quality of your sample is important. You get back what it hears. If you have a lot of noise it, will have that too in your song. So make sure to record it in the best way possible.
Can we get the amount of characters upgraded for Style of Music so we can be more detailed in what we are trying to create?
I'm having some bug, on mobile, where half of my screen gets black and I need to refresh page.
A way to regenerate a song to increase the quality, e.g. remove unwanted white noise and static, clean up sounds and the mix
But doesn't change the overall song too much
I love that you can use your voice to extend now, my suggestion to this is, can we use our voices as the another toggle option for a prompt maybe? Because when I input my voice and letting the AI know how my song's rhythm goes, it ends prematurely because of the extension feature. So my suggestion is to not extend the audio file but use it as the input itself, that would be a game changer. I tried like 50-60 songs yesterday and only 5 ended correctly through extension, even for a 6 seconds audio upload.
š Can we have a simple button on the player to have it playing in **random **order?
A feature that allows you to upload samples or songs that already exist to obtain prompts that would reflect how to make a similar one while avoiding copyright issues.
Or [Prompt] parts for inserting noises that exist in nature or traffic etc
or for sounds of different types at all
Would be nice to be able to generate like Bird noises and jungle ambience to add to a track in some editor
or rain
I wish there was a Star as a 'Favourite' button. "Thumbs up & Thumbs Down" are good but a Favourite button would help me distinguish the ones I really love.
Also I wish the stems were actually created individually in their own tracks in the first place and merged together to make a song. Right now i think it's making its job a lot harder to process and seperate them.
The ability to have songs as duets if specified in prompting
Option to just have a black background with white text for Video Lyrics or customize it with your own artwork.
If you are a paying customer the second thing you just asked for is a thing, uploading customized artwork
Yes, I'm on $30 plan. The second thing only applies to Cover artwork on the song, not the video.
Upload your art, finalize your lyrics and title, regenerate video
Ahhh ok. So that will update with the artwork. Thanks. Just no feature for the Black bg with white text.
You could probably get close uploading a blank background
But as of right now, no, no black and white box or text thing
Oh yeah. That makes sence. Good callout.
Thank you! Doesn't make sense to have that feature if you can technically already do it as you described. Thanks!!
Hey: Not even a little bit of a problem
An advanced feature of uploading audio where you can upload multiple files like a stem and the AI takes the data from the audio input, merges and generates the prompt given. Seem like a very far ahead thing, but thought I would share an idea.
I have a couple songs that are really good, but could be great if I could go in and just edit the lyrics to clean up some awkward originals that still came out solid, into something that fits the pattern better and hits harder.
There's another with an awkward ending that came about because it tried to end early with lyrics left to go, and stumbled through the rest. It had an AMAZING finish, but trying to cut out the awkwardness manually doesn't come out right.
I'd love to see a way to be able to fine-tune a song we've already 'finished'
?? What the.. I just said that I want the audio input feature extended so that I can upload 2-3 minute audios, I hate it when I have to stop abruptly at the 60 second mark when I'm recording guitar and such.. I want to essentially also see a feature where if you upload a file, you can get Suno to gather the other instruments around that instrument.
Could you make it generate less bass? Almost every song has extreme amounts of bass in it even when it shouldn't. And also, some sounds like the kicks and synths sound very plucky, and almost unpleasant.
Would be nice do have a "Delete all trash permanently" button š
Add the ability to just have a solo vocal, i.e. no backup vocals.
While you can't use negation I tried every possible tag suggested in the forums without success to have just a single solo vocal with no success.
WHY
In modern music composition many musicians process vocals with various AI singers and it cannot be done if backup vocals are used.
Thank you!
Yes. It adds some background noise to almost all my gothic doom metal songs, so it would be great to have an option to remove that.
i got a big one, when buying the pro plan for a month and ending it since you aren't sure how much you're going to really use it, not lose my remaining 880 credits I haven't used the second the month goes over would be very much appreciated. XD
The ability to filter by Full Song would be helpful.
when creating an custom, instrumental song. "style of music" is limited to 120 characters and suno's genre/style suggestion buttons. would be cool to enter our style/ideas with longer or no limit to the characters. it's difficult to convey creative music ideas/direction when limited to 120 characters. hope this is in the works?
the style of music box is for the overall sound and style of generation you get, you cant/shouldnt rlly guide it; but you can do that a bit in the lyrics box by using metatags (eg [fast-paced drums] or [Soaring Cinematic Strings] in the lyrics box) which can help guide suno in a specific direction but its not 100% guaranteed. the main style box is kept short bcos any more than that would be pulling in so many styles and training data and noise that you're going to end up with a generation that actually barely resembles anything you want at all. the less concise and precise you are, the more varied and inconsistent your results will be :)
when you figure out how to 'polish' the song, hopefully we can go back to previous songs, polish or pull the stems and polish. (Polish being remove the extra noise, make vocals sound more clear)
Suggestion - Hybridize creations.
Something were you can select 2 or more songs to have the Suno hybridize them together, mixing the two songs.
re-using seeds for voices, so when we extend, the vocals stay relatively the same unless implied different.
the vocals changing style, speed, inflection ate too much of my subscription
would also like to see v4 have a "sample" wheel, to create a better construction flow
the current "ringtone generator" is not an ideal format
the problem then is extending and re-using hooks doesn't preserve, or allow an inflection
this breaks any sort of consistency and control
would like to break it down better but not sure if it'll fall on deaf ears
Rewarding Tracks within a certain follow range
Being able to award creators with Credits, following a "tipping culture" Have the posibility to add 1,5,10 credits per track?
try [sample.speech] in your lyric title headers or [sample speech]
Hey everyone! Iāve been thinking about a few improvements that I believe would make Suno AI even more amazing and would encourage me to use the app more frequently. Iād love to hear what you all think about these ideas:
Adding Your Own Vocals: It would be awesome if we could add our own voice to the tracks instead of only using Sunoās voice. Imagine being able to create music with your own voice directly in the app!
Separating Music Elements: Another cool feature would be the ability to separate the generated music, like having the vocals separate from the beat, for example. This would give us more flexibility when creating and remixing tracks.
Improving Music Generation: Of course, improvements in the quality of music generation are always welcome, but I think the ability to use our own voice would be a huge game-changer.
I believe these changes could attract even more users to Suno AI, and Iād definitely use the app every time I wanted to create music. What do you all think of these ideas? Do you think it would be useful to have these features in the app?
Iād love to hear your opinions!
Add the ability to 'finish' a song with option left and right trim with fade in/out.
Add the ability to toggle if someone can extend off a 'finished' song.
here's one: some kind of daily credits for premium tiers. At present, you get a set amount right at the start, and that's not really much more than a free user per month (50* 30 == 1500, vs. 2500 monthly, only a 1K credit difference). That difference can be eaten up pretty quickly, especially given how rigid 3.5 tends to be at times on structure.
Yeah, I always found it kind of strange and unfair that premium users have not so much more credits than free users.
I mean, I'm out, and already blew through a reload, with two weeks to go
3.5 is way too hard-headed about structure
gonna repost this from the other channel
honestly, the biggest improvements v4 could make would be giving you better tools to look under the hood of your generations and see what it did (so you can more easily replicate what you liked), better tools to clean up songs you like but want to make truly shine without letting it go crazy with what you have already, and a bit more flexibility with structure
Are there any plans in the future to add any more controls to Suno? maybe i just dont know how, but there is no way to create consistent songs, with the same sort of vibe. Ive had some really unique songs that came out that i would love to make more of, but since it is very random how the song will sound i cant get a similar sounding song among different songs. if that make sense. maybe some control over the seed, or singers or replicating a quirk of the AI that are one offs. i think that would be amazing
have you tried extracting a piece of a song you made and then uploading that as the start of the next song you make?
i havent, but i will try that.
I create music using FL Studio and then upload it to Suno to experiment with different sounds. However, I'm not very familiar with various music genres, which makes it challenging for me to decide which genre to choose when extending a track. I wonder if Suno AI could identify the genre of the audio file I upload or even suggest a genre similar to the one in the uploaded track.
Thanks
Will you implement a feature to create original songs based on the melody and instrument of an uploaded audio file?
I think it'd be awesome to have the option to "save" a specific voice mechanic from created songs, with the ability to choose "saved voices" for new songs
If removing autoplay and repeat buttons from the mobile player interface has been done on purpose, then the suggestion is to put them back
It would be cool to have a commenting and replying feature in Suno. It would be nice if I can ask a tip on how their songs sound the way they do.
In Suno, does the separation of voice and instrumental work well or does it leave artifacts?
im back i have a suggestion why dont we have a mass down load button where we can down load all of the songs we made even in the bin the ones we threw away we could down load those too in mass
I really want the ability to work within the context of a project (see: any product of any type that allows for working on anything). Functionally this is essentially a special type of playlist that is kept separate from my other playlists because I might have hundreds of projects and only 5 playlist I want to be public. The workflow would look like this:
- When selecting custom also provide the ability to optionally select a project (somewhere in this new expanded UI)
- When the track is created it is automatically saved to this project
- The project should be persisted between tracks such that I don't have to find it each track
- If I navigate away completely from the creation flow then it should reset. The UI should also provide the ability to easily reset
- I should have the ability to search by project name and filter by creation date
š
I would love to see something like this
When you generate a new song that each generation has some sort of code for the music and singer so that you could regenerate again or generate another song using the same musical and vocal elements that had been previously generated.
So many times I want to make a few tweaks but end up losing a nice generated style.
I'd like an option to link certain Parts to other parts that weren't originally in the same song, so to speak.
I was able to get an amazing Parts 7-8, but have thoughts that it would be good to completely redo everything before them.
Unfortunately, I'll end up losing those amazing last parts then.
Please add a filter in the explore window.
When i put jazz i will see only all the styles having jazz in their name.
Please add 'Retry Payment' button somewhere under the 'Manage Subscription' page, as it's annoying AF to wait until your card gets charged, as I've been waiting for hours already
I just wanna pay for the fecking service and start using it again.
I didn't subscribe for a month, so my subscription is not activated now and I just can't get it activated also
finally, it worked!
Maybe add the feature to "extract" the instrumental part out of the already created song. Because when the "voice" does not sound good, but it is generally a nice track, we would be able to apply a nother voice onto it
The ability to crop a song, sometimes the end just keeps going so the built-in ability to chop it would work wonders
The feature exists : Get stems. Did you tried that?
/chirp
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
billing@suno.com š
thanks!!
any news when we will be able to use any other payment method then VISA or Mastercard? Paypal is easy to implement? That would be a great advantage to people who do not have a credit card.
I'm using lyrics from a historical text and I'm not allowed to use the Latin word for black, I've had to change lyrics that were written in the 12th century...
I don't want people making racist songs either, but they will whether you allow this word/sound or not. Disallowing this word/sound creates a functional barrier that should not exist. Please consider removing it.
Ā· folders
Ā· download all
Ā· web editor like FL Studio
Regarding Stems. It really needs to be 4 fold. Vocals, Instruments, Drums, and Bass. Instruments being melodic elements like guitars, piano etc etc. Thought I would be my 1 cent in hehehe.
a way to mix different songs together and let the AI spits out a new song from it. or you can take different part of other songs and let it generate and combine those parts together
a way to reuse voices and music
Yes, it would be great to have an opportunity to link any song or part of a song to a different song or part of a song.
Try iZotope Audio Editor for that. It has the Music Rebalance feature that can divide a music file into Vocal, Bass, Drums, and Other. I wonder why it has no separate Guitar, though.
Bandlabs also has pretty thorough stem separation and options, but i'm dont need those tools enough to spend the $$$ for the deluxe options
Progress bar for v4 development
I use to respond to stuff like this with a Soon joke, but it's been nearly three months since Tony told us it was cooking, so now you get Don't Hold Your Breathā¢
In-Painting | In-Painting | In-Painting | In-Painting | In-Painting | In-Painting | In-Painting | In-Painting | In-Painting | In-Painting | In-Painting | In-Painting | In-Painting | In-Painting | In-Painting |
.....and the ability to upload a whole song.
in-painting is coming with v4
Oh that's great news. Do you know when v4 will arrive?
Just look above where you asked for in-painting and read my last comment
That's disappointing.
It would be nice to at least get an update
indeed
It's been in development since at least March when v3 was launched. Not sure if there was any delay added due to v3.5 being released (which wasn't originally planned as far as I know.) Most likely in the later stages of development by now I'd expect.
Yeah that v3.5 came out of left field. I thought for sure that we were going to get v4 with everything I'd heard we were going to get.
v3.5 seems to have been created out of necessity given some of the issues v3 was suffering from (sound degradation, unwanted layering of vocals.) I'm glad it exists. I think people might be more impatient for v4 if we still only had v2 and v3 to choose from.
Don't you want to support the function that only remakes instrumental through seed value when performing stem broke? Instead of doing voice and broke like now,
I don't know if it's been suggested already; but it would be nice to have an option to increase the entire volume of the generated output; some of my songs are particularly quiet :p
I would love it if the browser version of Suno would have a better Library functions. For example the Explore section is more for song dynamics which is really confusing, most people would think it's to explore different song genres and artist on the plaform but it's not, it's a way to explore creative genres you haven't thought about. There should be one section dedicated to creativity like the library and create and explore functions. BUT! There should also be a section for searching, genre exploration, and just searching the database of really cool Suno songs, which get lost in the fact we don't know to search songs we don't know about, which is why genres and recommendations are so important, it's because people can search, but they can't search what they don't know.
I have a serious question. Can you please advise me on how to distinguish between works that are produced in the pro plan or free plan?
There is no difference when it comes to song generation. The suno model version listed next to songs can make a difference though. There are other post creation feature that make the difference too.
Suno moderation flag for nothing, it's useless asf, please remove that sh*t, ppl can do words phonetically anyway. It's just annoying to everyone, thanks !
I checked the results whether it's free or from the pro plan, it's v3.5 the same.
Hmm?
I take care of my community, which is serious about selling their work, but I can't take good care of it if I can't tell what work is produced on a pro account or a free account.
Im not understanding what you mean
Are you talking about when it refuses to do lyrics because of content? That's just ChatGPT refusing to deal with it, not really Suno's doing.
Please
1)increase the style Window's characters limit.
2) Display the total number of the pages in the library and give a mean to go directly to any page.
Hard to understand what you mean here but you cannot sell the songs produced by a free account. Is it that you are refering to?
For #2: You can type in a page number in the library, right on top of the number. At least in the Desktop Browser library, not sure about mobile.
Yeah. It works on mobile too. Only one character appears so you dont see really what you are typing but the feature is there. Not really usable on mobile. Thanks any way for the info.
Oh tavern works a little bit, but it's very weak.
I suggest an essential and easy to put tool for Suno: the buttons to go back and forward 10 seconds. Even 30 maybe
I dont know if anyone else said this but it would be nice with seed numbers for vocals, lyrics and instrumental. So you can start over when you write something in another language the ai cant pronounce or spelling mistakes
or if you are writing a song and you need a specific voice to match the lyrics
Marinating more then cooking š
Would be nice to have a notification added when we get a new follower on the Suno profile page. this way I know who to follow back without searching 100 names.
make it possible so that you can continue the Generated lyrics instead of going from the last verse or chorus
Some "artist" filter words are utterly ridiculous. "Fuego" is a word, just one, and it is used on a huge number of occasions, especially in songs, without referring to the artist, but Suno blocks the song indicating that it is a name.
All kinds of censoring to a creation tool for music is ridiculous, all it does is force serious users to move to local ai once its there and it will be. It is already here for text and images. And people who wish to use Suno or others for racism etc can just work around it. It is especially annoying to users in countries that don't live in "freedom" under the FCC. I tried making Suno write a folk-song with only curse-words and all it did was give me 20 times goddamn.. lol
could provide a way to search for songs that have been deleted - a search field in the trash.
i need a option to download a whole playlist or maybe multiple songs at once
Any hints at the future of Suno? I've been loving this for months but someone recommended Donna and it has the controls I need, singer gender (boy or girl) and unlimited song description length! I'd prefer stay with suno though, I just need those features added soon. For simplicity I would recommend the same song description field allow for also describing the singer/s features freely and it obey. Right now it completely ignores us about the singer.
Try male/female vocals, I have found that to be pretty reliable.
The ability to sort songs in playlists on mobile would be nice. I have several lists I curate that I just add things to and have to remember to cull and sort later on once I can get on my PC.
Would be great if we could download the .srt file, so we could easily add subtitles in our own videos.
At least another 20 characters for "Style of Music" At least!
As much as I'd like to take advantage of it (in the future, I did it this time to see if it were even possible out of boredum). The system isn't accurately tracking credits when you are on your last ten credits. This means for your last ten credits I can spam click create and for the price of 10 credits generate as many songs as the queue gives me capacity.
Clear Trash button
i don't think if anyone suggested this before but
why not increase the duration limit of the audio?
i honestly think that sometimes it isn't a good idea having the limit of 60 seconds per audio
Can we have a way to view all the songs (our own and other user's songs) we have liked?
The Liked filter in Library only show my own songs, and the Liked Songs playlist doesn't appear to show all the songs I have liked - currently appears to only be showing approximately 50 songs (is that a bug or is it only intended to show a limited number?) when I have liked more than that.
Is trimming audio going to be coming soon? I have a number of songs that have mild hallucinations at the end that I'd like to cut off.
I've seen a button to exclude styles that made another field appear to type styles to exclude. It's been briefly available a few hours ago but is now gone. What was that?
I wish we could extend an uploaded audio without specifying a style and let Suno do its thing.
Add crop tool to edit length of files to upload. Most of the files I try to upload get rejected cause they are too large.
Add "favorites" and/or "recently used" under style of music. Similar to the suggestions we already have
Stems: We need detailed stem separation, each individual instruments as separate track, and male, female and chorus as separate track if the song has many types of voices. Later its easy for us to control the volume.
Right now Vocal stem had little Instrument noise and Instrument still had bit of vocal in it. please refine stem separation. Which is very useful.
Some time the song generation is not correct, Say some time no lyrics in it, complete noise. Such tracks should be reviewed and the credits should be refunded.
And one Important suggestion is need consistent Voices, If we want a same voice for different song creation,
if any option to connect, or save the voice as preset for us to use for many songs, It would be great.
That's the only option for stems from the AI, given how it works
It's not creating stems in the first place, and the quality of stem-rippers is only so good.
When will you give NSFW creators an avenue? You could require them to mark songs as NSFW instead of forcing a moderation system on people.
Can you add a text input for Bugs Other to add an explanation.
Searching on tags for some styles don't return results any more: goombay, gumbe, taarab. Meanwhile other tags I've used like ) - is it pizza Friday? return results. This lack of results is true even when changing the different filters in the drop down. Previously this would return results. Given Suno displays just a black screen when nothing is returned this feels more like an error state (versus displaying: 'no results found'). I observe this lack of gumbe on both Brave and Chrome, but where I feel it most.... where I feel it most... is in my heart.
can you guys add a feature to ur delete account?, in some countries it is illegal to not have a option delete ur account.
the way of emailing the company in order to delete the account is getting complicated
Definitely, this is something I am missing. Also, better filtering, for example being able to filter only disliked tracks.
Add the ability to just have a solo vocal, i.e. no backup vocals.
While you can't use negation I tried every possible tag suggested in the forums without success to have just a single solo vocal with no success.
WHY
In modern music composition many musicians process vocals with various AI singers and it cannot be done if backup vocals are used.
Thank you!
I would like a library of public domain MIDI files that can guide suno, so that suno can create new adaptations of public domain works.
The public domain files should be selected by suno for legality. For example, I would love to have a rock adaptation of the Blue Danube, and of Beethoven's 5th.
The public domain exists for this purpose, to let old works get new life.
Yes, it is gone now. It was handy
Add an "enhance"/"improve" tool which just enhances the audio quality of an uploaded file. For example if some sound was recorded on a phone and it just does not sound like studio quality, then suno should enhance it and make it sound like studio quality.
how to just sing no song i want to make different language cover for a song and i cant make it with no music
how about a comment section for each song? like on the songs page itself? sometimes i feel like expressing my feelings for a song through a like just ain't enough. Maybe even a DM (Direct Message) section?
didnt they just come out with stems a couple weeks ago?
The Suno follower count is off by 1 when viewing the profile page. The API: https://studio-api.suno.ai/api/profiles/followers?page=1 is returning the correct number in num_total_profiles 50. The correct number are displayed on the "Followers" page 50. However on the profile page it shows 51.
When I like or unlike a song it also doesn't change the like count. This makes me wonder if this page is cached and isn't being refreshed when these metrics change which is odd given the API seems to be pulling in accurate real time data.
`
Please have the option to make new playlists appear at the top
OR
Last used playlists show at the top.
It takes a long time to scroll all the way to the bottom every time I want to add a song to a new playlist when adding new songs.
I'm on page 500 of my music going back to v2. trying to sort everything into folders is really time consuming (thank you for adding my idea of typing the page #)
Sad we cant Get Whole Song of v1 and v0 stuff from Discord and there is no correlation markers for each part. The videos that were generated (ā¤ļø) within Discord: I don't see those recognized as a Full Song in our Library
I'll throw this here. Suno appears to be "leaking" meta data for all songs created by other users. Using this information I can see all extensions for all public songs.
I refer to this as "leaking" because I as another user have no reason to have access to this information, and it isn't being shown any where in the UI. While it's not sensitive information, it does let me see how many extensions were required for a given song which is sort of interesting trivia given the UI doesn't expose this information. There's other tidbits here that I can't yet figure out how it might be useful to me yet.
This information is available through any API returning a clip object. If it were my backend I'd remove all information that a user didn't need to know about. It also would reduce a ton of network traffic.
Also with Metadata, the Creation date is instead the Download Date, which doesn't match real creation of the song. š¦ š¦
make a new channel here called "trouble-shooting"...and also a channel "suggestions-audio"
A) ability to save style and text prompts
B) prepare a library with genres and subgenres for musical styles, as well as the possibility of varying the intensity of their presence;
C) as above but referring to other characteristics: atmosphere, environment (live, club, room, recorder...), and above all styles of VOICE with the possibility of duets, choirs etc
D) set up a guided structure where for each part chosen by the user (which could also be totally custom as it is currently); within the structure give the possibility to insert both music style and text.
What the heck Alex Ayers has a real name
and that name is
Albert Einstein
I have suggested this because it makes it difficult for me to stop playing an instrument at 60 seconds.
it is so difficult to implement a cutting tool to cut the songs?????
it is not, but Suno is too lazy and don't care about the users...
this is a 1-2 day job for real programmers.
ā Warning logged for grabby_35252. I couldn't DM them. || Please refrain from making disrespectful comments. Constructive feedback is welcome, but name-calling violates our community rules.
@clever harness I will have to remove your post since the files do not have an extension and therefore could seem malicious.
oh my bad let me reupload and fix
Thanks!
I think the most annoying visual issue I see on the browser (I'm a MacBook pro 16") is the flexbox on the following and followers breaks starting at exactly 1536 pixels in width. I see this everyday, because for whatever reason this is my preferred browser window size. Notice in screenshot two it's prematurely loading in the column of images. This is only happening when you get to column 5.
can we give suno some music and create a song by just composing it ourselves? how can i do that?
in terms of having fine control of your generations (like maybe giving it a midi or specific passages of music to generate), there isnt that level of control currently - it's just prompt-based. what you can do currently though is upload some of your own audio and have suno extend from it, which will use similar sounds and melodies from it alongside your style prompting. there's an Upload Audio button in the top of the generation screen.
It would be beneficial if Suno could incorporate a feature that allows users to input a song or file URL and automatically identify its genres.
can we add our own bg music for the lyrics we give, and the lyrics the ai will sing adjusts to our bg music ?
Would love to mass delete (or an easy way to delete) songs I have made and given a š .
I'd also love increased filtering for the library, like being able to search for specific date ranges, songs with a thumbs down, or being able to see all "get whole song" results.
While it's only a consolation prize, we ARE able to delete multiple songs at once using standard desktop selection options, like I show here -- hopefully this can make library management a little less tedious (a little)
https://cdn.discordapp.com/attachments/1270928841614823546/1270928991947063386/multiple_song_selection.mp4?ex=66c0b11c&is=66bf5f9c&hm=691b2c455d0763ae583b8fa3e6b93e938a00f4cefef49944a75483c3de9097a7&
Holy!!!! I don't know why I never tried this š¤¦āāļø
Add the ability to just have a solo vocal, i.e. no backup vocals.
While you can't use negation I tried every possible tag suggested in the forums without success to have just a single solo vocal with no success.
WHY
In modern music composition many musicians process vocals with various AI singers and it cannot be done if backup vocals are used.
Thank you!
The library list display should be changed...
1- Multiple selections should be achieved with a checkbox (per item)
2- Add Folder function, with the ability to move multiple files into them... Would be very useful for management
3- The items should not be displayed as page, but as a long list we can move up n down using the mouse scroll or a scrollbar...
4- Every Generated file should have an incremental number between ( ) after the name...
test song (13)
test song (14)... etc
When we generate a song, some randomizations are used (that is normal) so each time we press CREATE, 2 new versions are generated... but those seeds/random numbers should be available after the song is generated... Why? user would have the possibility to use some or all the same from a previous generation, If he just want to change some lyrics, and keep the rest identical... Instead of spending tons of credits trying to get something similar as what it was (I know there is a few workarounds for now) but this is how it should be.... But with the amounts of credits spent by users chasing another Generation similar, Suno is not in a rush to implement that last idea...
Of course the followings can be achieve outside SUNO, with 3rd party applications on the WEB or on Desktop.. but once you have done this process, the song cannot be brought back into SUNO !!! (or just a small part as UPLOADED audio)
--> Maybe these last ones should be available for all users, but for a new SUNO PRO subscribtion... And all these functions are easy to add, most of them are already available as JS library ...
A simple audio editor, (similar to Audacity) should be added, with a stem separator with parameters...some easy to use preset for filters like REVERB for the vocal track, Compressor/limiter/Eq.. for mastering the song...
A good example is that editor https://audiomass.co/ just 65KB of javascript executed locally...
Enough for now...
.... Useless to dream ....
ā¤ļø Dude -- this is a phenomenal write up š š
useless to dream
Not in the SLIGHTEST šŖ
As far as I know, the devs poke their heads into the bugs & suggestions channels to look at ideas & see what people want from Suno, or what they should improve first, so seeing technical and detailed feedback, along with you reasoning behind it, probably goes a long way to help get them talking about it at the office, instead of people making similar ideas (but somehow making it sound like an angry complaint rather than a thoughtful suggestion)
Not sure if it has been mentioned before but adding a button for full songs next to the liked and public ones would be nice as you could go there and see what songs you've got that are full ones. I generally don't like to hit Public on my songs unless I have made a corrosponding video to post on my YouTube channel which is where the Full Songs button would be nice so my library page would only show the songs I've finished. I know making a playlist with all the finished songs would work too but just having a button for full songs like the public and liked ones would make things so much easier in case I forgot to add the full song to the playlist, it would still appear in a group.
How about some Reaction Emoji's for our songs?
I don't understand why there are any restrictions on prompts whatsoever - it seems to me that it is just a matter of time until we have a censorship free service and it will be the best service for this reason
lol
the level of moderation and censorship here at Suno is way too intense
i'll be switching services to a less restrictive service as soon as i can
Part of the work process uses ChatGPT which rejects content on it's own. A lot of the "censorship" is just GPT not being allowed to do things. You can often try phonetic spellings to get around stuff. It has been quite funny to find what it censors sometimes. "Girlfriend" being rejected but not "girl friend" for example.
the quality of the voice melody generated by V3 is WAY better than V3.5 in almost any genre. There are worlds between V.3 and V3.5! It seems to me that V3.5 has been trained on stupid EDM songs and mainstream "cheap" music so it generates also "cheap" and easy melodies. If you want to make good melodies switch back to V3.0. I hope V4.0 will include somehow the V3 voice model or at least an option to generate longer songs with the V3.0 genial voice generation.
Am I the only one who loves Suno but is absolutely bamboozled by how bad the front page is? Imagine going to Youtube and instead of videos that you like they show you the same curated, non-randomised, videos for months. Youtube would be dead by now, but somehow Suno thinks it's a great idea? I've been on Suno for roughly two months now and still I see the literal same songs in the same order EVERY single time I open the site. A summer playlist without summer-style music and another showcase. Is Suno working on some kind of algorithm that shows you songs based on music styles you've listened to? Or at least place the trending songs above? I unironically feel like Suno has the worst front page of any website I've used in years.
As I assume already for 3 months, the UI programmer is dead long time and they have no replacement for him...
I think I noticed this too.
Don't be disrespectful. You have been warned about having this attitude towards the developers before. It is not constructive. Post suggestions about what you'd like to see improved.
There has been UI development within the last 3 months. The entire look of the website changed. You can now go to a specific page in the library (something you asked for repeatedly!) Audio upload was added (which required UI work.) There are several things currently available to alpha testers that involve UI changes. Just because you do not see things changing on the website does not mean that nobody is developing the UI.
The Similar naming works perfectly, me I usually double a letter, "ii" or invert 2 letters "ei" instead of "ie", as long as there is a TYPO, the discriminator, is not part of an AI model, it is a pre-pass through a simple list of restricted names... once this is done, a small typo wont prevent the model to figure out what you wanted to write...
I even ask sometimes for a fusion between the style of "XXX" and the style of "YYY", and it actually create a new style inspired by the 2 artists !!!
I'll keep my frustration in check and remain polite, but this one is truly a baffling decision.
When a standard is well-established for so many years, it's unwise to try to change it. Take copy and paste, for example - CTRL-C and CTRL-V are untouchable functions and shortcuts. A programmer who dared to change them in an application or use them for something else would be immediately stripped of their diploma and publicly shamed!
Keeping that in mind, let me point out a fact: all musical works we humans have recorded, on any medium intended for public release, share one thing in common: the "sticker/cover/image" has a 1:1 ratio. In other words, it's square(ref. Fig no.1). From the 78 rpm vinyl records of the early 20th century to the compact discs of the digital age, and even the humble 8-track cartridge and the ubiquitous MP3 with its ID3 tag, the square format has reigned supreme(ref. Fig no.1).
The rectangular 4-track tape case was an anomaly that caused countless problems. We're no longer in that era, and we usually learn from our mistakes...
So, why on earth did someone at Suno decide to change the already perfect square to a vertical rectangle for displaying album covers? Find me one place where a non-square shape is used to show a musical work cover...
Please, please, change it back to a square. Many covers are currently missing parts in the display. For videos, a rectangle is fine, but the image itself is still a square (ref. Fig no.1) with the lyrics underneath. For genres, artists, or folders, use whatever format you like, but for a song or album cover...
IT'S SQUARE ONLY... (ref. Fig no.1)
Sorry for the insistence, but I'm confident this message will reach the right people.
And I feel better now... Better laugh than yell !!!!!
Below is FIGURE No.1 : this is a SQUARE....
One suggestion I haven't seen, but what I'd LOVE to have is the fact that 60 seconds long upload limit is too short
aka, I wanna upload a 2 minute stuff, that has more arrangement and context regarding how I want the track to play out
right now, I feel that 1 minute is limiting me
Make the ai reference to a audio clip
I'd like to see extend it from the behind the audio clip, if you know what im saying
not dirty, I promise
I was just thinking about that. If you could do that, maybe you could pick a hole in the song "extend" it from both sides at the same time.
I'm not helping am I? š
I would like to suggest a remaster/cover feature, that lets us use a previously generated (or uploaded) clip as a basis for a direct remake of it with a different genre or altered lyrics. Essentially a song version of img2img. Being able to introduce song influence from uploaded/referenced clips without directly extending those clips would also be very helpful.
A mashup/hybridize option would also be neat, let us combine multiple clips directly into remixes or new songs with blended aesthetics.
We need a way to clip tracks. Often a song will end, go to silence and I think, "Yes! Great!", but then it starts the song again... so frustrating!
It is. And you probably already know this, but just in case you don't...it is possible to resume from a silent point like the one you mentioned and extend from time there, using the tage [End] and it can result in the silence you want. Not quite as good as clipping might be but possibly a workaround until such a feature could become available if one were being worked on.
Unfortunately, that rarely works.
As the vocals can now be stemmed out, why can't the other elements? That'd be great option to have too
please allow option to custom set end of song by simple trimming from a time stamp within the site. after hours of work I'll almost exactly what I need, but need to export and post-process to trim the end. if u add this feature I can share more native tracks without leaving the platform that are up to the standard that I deem worth sharing
eleven labs has some robust abilities that allow the user to tweak the vocal output that I think would be a great advanced feature set for suno as well.
variability vs reliability in the delivery, who much the vocals embellsh, the pace and such - there is so much untapped potential with vocal control that I am almost certain is in our scope, so consider this one point toward those features
I'm not sure if this has been suggested, I wouldn't doubt it has been, but i'll place it here anyway. Something similar to inpainting for image generation except for songs.
Ex: You have a song mispronounce a word, or forget part of the word or even part of a whole verse, but you enjoy the overall structure of the song and only want to change that part. So if it doesn't say something correct, you can put a timestamp between where that line starts and one where it ends and you can generate a new clip to fit that part. I'm not quite sure this is possible with your current setup, but from what i've seen of 'Get Whole Song' and how it can stitch together extensions, I think you might be able to implement that somewhat differently to get this, but I'm just making baseless assumptions.
Another potential way to maybe do this is let us mix-and-match already generated stems of the same song. Ex: I have a 4 minute song and I extend it around the two minute mark, but I like the later part of the 4 minute song, so choose the part of the extend that I want, then re-attach the later part of the song to it, with it making it a more natural sounding transition(first one is a way more powerful tool).
I feel like this is one of the big things i'd love to have added, since I usually get an incredible generation but something in there is wrong, or a whole verse randomly repeats, and to fix it I get rid of so many parts that I loved and that made it unique to make that happen. Also, I don't mind if each of these precise gens costs the same as a normal gen credits-wise. That amount of control would be worth the cost.
A Desktop app for at least listening to music. As a content creator, and someone who is trying to get more streamers into using AI music on streams as a better alt to DMCA free music. A lot of creators like the options for independent players for their music.
Suggestion: copy a style from a song you created.
Explanation:
Say I generate a really good generation, but it's not quite right, but nothing else has been this exact style. I could then copy that song's exact style, and vary off of that to further finetune what I really actually want.
You are, u for real just spelled out clearly what I was trying to say š
Suggestion: Punch in /Punch out generation.
Say you've generated a 3.5min song using the v3.5 model, having the way to, similar to Extend, generate new content between a start (punch in) and a finish (punch out) time and then blend back to the original generated piece. Use case would be to replace a verse or bridge that maybe didn't hit the mark, especially when not using the custom mode.
Suggestion: Use same voice. That way you can create multiple songs by the same AI artist if you will.
Suggestion: Ability to describe the song, not just the lyrics or style. IE starts slow then rises to a fast pace.
I'm wondering if this is the only place to provide suggestions to the Suno team? I have some ideas about the listening side of the Suno experience that I believe would make Suno a better listening experience which will also make things better for creators who want to be heard.
I'm not looking to complain - I want to offer some insights from years of software design to better things for the creators, listeners, and Suno as a whole.
Based on my own web application dev experience, I think several of the suggestions are minor tweaks that will create huge upside with minimal effort.
@foggy: if you read the suggestions from 3-4 months ago you will notice than none has been implemented.
yeah ... that is my concern. This format is so unstructured. I run dev teams - feedback needs to be more disciplined in giving and receiving.
I know they do the outreach calls - maybe that's a better funnel for ideas. Not sure.
I also see a lot of people grinding axes in here which, imo, is not constructive for anyone. We have to assume the team has a massive list of priorities that they are juggling and they want to make quality product.
if they need 3 months only to add a direct access to a library page you can imagine how the rest is... and you can imagine who programmed that to forget about a direct access to a page!
I wait for the day a UI programmer is hired who takes care only of the UI and front end.... they still have the add on their site, apply for a job š
Todayās top 30 Suno jobs. Leverage your professional network, and get hired. New Suno jobs added daily.
Be the change you want to see in the world.
it is weird that you all here offer me programming jobs every time I talk about the missing UI programmer and the missing functions š The whole UI could have been done or even redone remotely in 4 weeks with all the tweaks by 1-2 fivver programmers and a good project manager and a tester from your side... and you cannot do this in 4 months ... there is no big deal to find a programmer for the front end... and is not a big deal to programm it either. Man, in this branch people work remote... why do you still insist to have a programmer there when you cannot get one ? Just too bad, since the composer is really a genius ! So congrats for that, but the UI... oh man...
i wish i could use seeds from earlier songs i made to have something even more similare to that. as even within using the same inputs you sometimes get something that is hard to reproduce no matter what you do
Every time i read something from you its highly insulting the people making all this here possible in the first place. Whats the deal with that? It hurts even me and i am not even working for suno...
"insulting" ??? you must be dreaming... "insulting" is for you when one complains about the missing UI programmer or the missing functions any other programm has?
I made lots of suggestions, even drawings, in the past 3 months. I did a lot of constructive work, but in vain. I offered myself as a project manager for the UI stuff, also in vain. I also suggested to make user donations to pay a programmer since they don't have one that works on UI. There must be a guy there who works at everything and every once in 2-3 months he can do something. As a dev myself, I know how it goes. So we must probably wait I guess some 6-12 months to have some basic things done other companies make in days... and this is a fact not daydreaming.
tact is becoming more and more rare in this internet age,
On several occasions you have suggested that the UI programmer is dead. I would consider this insulting. How would you feel if you were developing software and looking at suggestions to see that someone had suggested you must be dead. Is that not rude?
Criticism of missing features is reasonable.
Making personal comments against the developers like this though is not acceptable.
never heard of sarcasm or humor ? maybe you take it for "insulting". It is a funny way to say that "there is no one there to take care of the UI". Is this insulting ???? Come on ! Never heard of humor ? It is more than obvious not only by the months needed to implement a function made in 2-3 hours by others that there no dedicated UI programmer there ... you can even read it on the job page š So I was right the whole time assuming there is no one there dedicated.... I wonder only when this will change! So why you all cannot you take any critics when this what I am talking about is a fact?
Suggesting that someone is dead just isn't funny. At all. No matter how many times you say it (this is not the first time.) Maybe you mean it in a humorous way but it doesn't come across that way.
Let people's points stack! Lots of generative AI sites do this, and I think it would be pretty cool if Suno did too.
Please name ten. Okay five. Can you name five?
Iāll start you off. Last i checked Kaiber.ai did this.
thats quite the interesting response
I am not aware of any others.
Nightcafe to add to that
I use AI for my stuff, a good deal is a good deal. Iām genuinely interested to know.
Will look into Nightcafe.
Can you name three more?
Because if ālots of AI services do itā you should be able to name that many easy. š
lol. sounds a lot like one of those people that go "wow you like [insert band here]? name [insert random number of songs here]"
real š¤ āļømoment
Nuh uh uhā¦now whoās trying to shift the goalposts?
lol still wouldn't be me
Your words.
taps the sign
Maybe not actually as common as you might wish,eh?
and? what bearing does that have on the point?
seriously i would genuinely like to know your thought process here. what part of the idea are you challenging?
because if its just the "a lot of" part then it really is a š¤āļømoment that does nothing at all than go "well you cant name [x ammount] of sites so hah!" without any real attention towards weather its a good idea or not
Not a fan of no credits stacking, I have used over 50 credits for a singular song and now if I run into issues where I have to continually use credits I will completely lose a good song because if I start it up again the next day it will start doing the random singing numbers/seed bug on the extensions⦠Not to mention completely losing āthe flowā when working. Please add an option for credit stacking. I see there is an option to contact and set up custom credits, but I canāt imagine that is scalable if being handled individually for thousands of users. Iām also unsure now if I am going to run into issues when having credits from the previous subscription system and then canceling Premiere to use up the credits I have versus continually paying $30 dollars every month for 50 credits a day I will never see. I will be reaching out to Customer support when the time comes but I just donāt see 50 credits being enough in a single day especially when you are in the flow and working all day on songs.
Edit: Personally more like 500-3000 credits a day because the majority of my time is lyric writing/ideas then getting them done 1-2 days out of the week once I have a clear idea/genre composition in mind versus āgenerate random song once a dayāā¦)
i feel like they've already made a step towards stacking anyways with their change for pro/premiere users getting the 50/day after running out. would be nice that in general that 50/day could stack day after day. Same for the pro/premiere users, or users paying for custom ammounts.
Another idea I saw Nightcafe use: small bonuses for certain activities, like daily posting, sharing, getting certain ammount of likes/shares, etc
Your humor is not funny sir. Really. Like really really unfunny and also a bit annoying to be fair. sorry
It is not humor. It is sarcasm about a sad fact.
ok. sorry. then its that.
"never heard of humor?"
"its not humor"
pick one?
Quote right if you quote, dont cut out what you want, I wrote: "never heard of sarcasm or humor?"
lol
sarcasm is a special form of humor, is not a joke š
i thought it wasn't humor š¤
I dont even know what to say anymore without sinking on a similar level. I can just suggest to rethink some of the ways to interact with others.
also I did quote right
so what ? should I tell you again that sarcasm is a special way of humor?
to be more precisely
i think you should make up your mind on whether or not it even is humor first
i think you should mind your business!
lol nah
I was not talking to ya
and the other ones are grown up and dont need defense from you
š¤
My business is ... here.
I was talking to Kilonova Mr. Tree
i now have to close this tab before going insane
I go sleep, bye
Oh!⦠Do we still get the 10,000 credits and THEN if we run out it gives out 50 unstackable credits?ā¦. I was thinking the post meant no stacking whatsoever so we could only have 50 credits at a given time. Belay my last if that is the case.
ooh good question. I think as of now its no stacking whatsoever though im not sure how paid monthly credits work. do those stack? like if you dont use all your credits by the end of your monthly pay period, do you only receive enough credits to get you back up to your respective paid ammount? I know for the 50 daily it does not stack period, and youll only recieve enough to get you back up to 50
i would love to see it change to stack in general.
Ok, if we are still getting 10K a month Iām happy, I donāt need to keep my sub running and then get 20K, 30K credits if I am not actively making songs each month
true but would still be nice to recieve that ammount monthly with no cap since thats what you are paying for
yea⦠maybe like⦠get a diminished return
i do think the stacking would be more useful for those on the free plan. 50 credits doesnt get you a lot of songs, and i know I for sure dont make songs daily.
10K, and if you go to 5K, you can get 15K max the following month, but if you are at 10K the 3rd month it would drop to like 2500 credits on top of that
Or something idk lol
im sure if they chose to allow stacking the devs would find a nice and fair way to balance it
would be nice to have a new creation be made to be similar to an old creation.
like sometimes you might use the same lyrics, same style of music, but just not get what you want, even though something you had before was something you really liked. I've got one where I'd just like to change it from male vocals to female vocals, and any other minor differences would likely be fine - but there's no way to get close to an older style without the very broad "style of music"
Among many things, I tell you what is needed the most, from a heavy user point of view:
- Make finally the playing song visible ! (the animation is not enough to be able to see it fast)
- Make the most used functions as download and get whole song directly visible in the song line (not via drop down menu, this saves a lot of time)
- Add keyboard shortkeys for next song, previous song while playing.
- Refine the search function (right now it looks even in the lyrics for words, but would be fine to select "only in title" etc
- The possibility to add an intro (UDIO can do this) to any song made
- The possibility to cut the song end (audio editor online)
- The possibility to rate a song with 1-3 stars.
- The possibility to download many songs at a time (wav or mp3)
- Add folders to sort the songs by dropping them on the left side (like in Spotify)
- Negative prompts
- Forcing prompts (dont tell me this is not possible, it is!, Udio can and any AI war drone can do this, a robot is not a CAT to do what it wants)
- Shuffle function for the player
Like everyone else saying. Need shuffle button for playlist playback!
I'd love to see a character limit extension for Genres, When I can't refine my thought into simple genres; playing around with sentence structure, word choice, and positional contexts to shift the perspective Suno interprets.
I don't know what others feel like, but a boost to 160 from 120 feels like a reasonable amount
(It is the genre section after all)
Anyway, had that on the mind the last few weeks so I wanted to air my thought
I recommend two simple stuffs into the UI on the** library page**:
- bulk selection in the library, so we can select several musics and add to a playlist or remove at once.
- re-ordering the list in the library, so we can see the old ones first or the new ones, like the one we have on the profile page.
And again I would request having a š random button to play musics in random order in the player.
Thanks š
Why would you need a limit from Suno? most people want the opposite!
I mean, if they're willing to remove limits over all that'd be cool, but I'm not going to outright ask for that with genre section because I'd prefer the lyric section to get that treatment
Select songs using control+click or shift+click like Windows Explorer to select multiple.
Thanks!
It currently doesn't repopulate the page properly when you remove multiple, and that stops you from going to the next page, since it thinks the current one isn't full. Easy enough to fix by moving to another section and back or refreshing, but I'm hoping they fix that.
is there a way to share your creations here?
That's what the #š£āsuno-showcase channel is for.
Hi guys. I REALLY want a feature where music in the player can be downloaded to a predetermined location like Spotify so that I don't have to redownload all the songs every time I want to listen. If I've got 100 songs in a playlist and download them in advance, I really don't want to download them again every time I listen to them. Is this possible on web/desktop or web/android?
Че как
preach
Please consider this idea for UI/UX instead of a system of folders
right now playlists have to act as both public catalog and also user sorting system. Ideally playlists are for personal collections of finished songs only for public or private use, and not user sorting/filtering
I really could use a way to bulk filter my gens for a song session
Make a set of chats in a side bar (like ChatGPT) that serve as project boards. All of the tracks a user generates within this UI now lives in this project.
This replaces a never-ending book of generations (I'm over 5000) that have no order besides names or playlists
I am looking for a way to start a new song and have all those generations exist in their own workspace. A good song takes 30-50 generations (a bad song over 100 lol) until it feels 'complete'. Sometimes you get lucky early as it is never the same process but the point remains: Sorting by project is an easy way to filter your generations from each session into its own chat/section of your profile.
Chat is one design approach, tag filtering is another. Hmu if you need anything clarified.
Allow user to extend from a time with a +- variance. example "Extend from 2:30 +-5" could be from 2:25, 2:27, 2:33, etc...
Also, "extended from XX:XX" in the meta data would be nice..
i mean for my android, it just saves it to the same place as the rest of my music, and is accessible by apps like YT music.
Please get rid of the rain stick noise, its in nearly every song i make and its become frustrating to hear. Sounds like running water or someone going SHHHHH.
Ability to edit lyrics in the same song, same voice same everything and have it rework the same song with the updated lyrics. Rather than having to create a whole new song that sounds totally different if you made an error in your lyrics (which I do often), as it helps to hear the song and then update the lyrics
I'd really like it if the style of music letter cap could be increased. I've found a lot of issues in my time using Suno where I want to add something to my prompt, but need to delete something else to fit it in, where deleting it could lead to the entire song sounding bad
bro we need a Duet or Musical mode so than we can make better duets or songs for a musical
V3 makes a o lot of time duets, even if you dont ask for...
i like this post #1252773477157699664 but why only end so i think it should be option for both sides of song especialy if we use own sample to generate music
what happened to the discord feedback suggestion channel? this discord is self-enclosed which is nonsense... all these people spamming the suno-showcase that no one is listening to, give us another outlet....
singer seed value so you can re-use the exact same voice in future songs
in my humble opinion, i'd love to be able to download some of this stuff at a price of credits or like 99 cents a pop
