#💡┃suggestions-web
1 messages · Page 19 of 1
yeah sure and when is the last time you checked new section exepct when you publish something
Trending helps people and there is 50 spots in there if any of your songs didnt make it in there then it probly blocked by filters
i get it. we all want to trend. I understand that. but cmon, you must at least AGREE that the word TRENDING doesn't make sense if you have zero activity?
if your activity drops then you slowly drop down on the list

I would say one of the MAJOR things that needs to be worked on with Suno, is the audio quality. Instruments and vocals are often clipped, overcompressed or just missing audio data. As an audio engineer, I have deconstructed Suno-generated tracks using various tools and noticed real issues with the generated audio. I know AI audio generation is still in its infancy. Still, the thing that will propel Suno to the next level will be audio quality (not to mention responding to prompts/lyrics/queues better). People are generating great songs, but they are far from anything that you could use for anything tangible (even with commercial licensing included in our premium / pro plans). Can these suggestions be implemented in the next release?
It's pretty hard to drop in activity from none. How would one have negative activity?
Isn’t deconstructing it against tos? 🗿
if others gain it ?
So, if your song has zero listens and any other song has even one listen, it should be higher in the ranking that yours?
Do you really not see how that would suggest that a song with zero activity shouldn't be in the trending section?
Since we have playlists, could there be a functionality for creators to make albums? It's not super necessary in my opinion but it could be a QOL feature
BASIC FEATURES ARE STILL MISSING IN THE USER INTERFACE
- After 2 months of suggestions and complaing we still need a direct page selection (right now is even a bug there, the "programmer" forgot to check if you click more "next page" button than the total number of pages you have!)
- We need a feature to be able to give a note/comment to a song so you know later what you were doing first time etc.
- We should be able to give something like 1-3 stars to a song, to be able to differentiate between the best versions. Only a like button is not enough!
- Add a direct access symbol for both wav and mp3 download, the share link and get whole song in that line where other symbols are, so we don't spend so much time clicking and looking in that slow menu idea. There is plenty of space there to put everything users use constantly.
- add a "next/previous page" button (or even a search function etc) also the the bottom of the window, otherwise we make miles of long mouse movements every day
- add a 1-3 stars individual rating for a song, the like button is not enough, so we can differentiate between the best versions of a song.
And also the possibility to display only 3 star songs, or only 2 stars songs, or all songs marked with stars etc.
- add a template with user favorite meta tags or a changeable default template so we don't have to put manually meta tags to every song (tags like "[Instrumental intro], [instrumental solo] etc . Because who wants a song with tags as in the actual default mode, with no solo and intro ? I guess nobody !
So why don't use the computer instead of the fingers in the AI age ! The DOS age is gone! - Expand the search function! It is amazing how limited the search function has been implemented, and this even only after user complains of the missing of such a basic function! Please expand to this: 1. look only in the titles, look only in the lyrics or look in both. 2. search also for partial words like "amazi" instead of only "amazing". A misspelled title can be never found with the actual function! Who had this idea to look only for whole words ????
- Eventually give to possibility to reverse the display order in the library from top up to bottom up (in case we look for some older songs)
- Like in the splice app, make a local saving folder for the songs and give also the possibility to download all songs at once that has been made in a certain day. How much time we spend daily clicking to download every song manually!
- Change that low quality 192 kbit to at least 256 kbit or let the user decide the bit rate of the mp3 file in the settings (there is no "settings" now).
- Give the possibility to manually reshape the width of the columns in the app like most apps do. With a full screen, names of the songs are left and the often used functions are right. With a wide screen that makes a lot of head movement every day! It is not ergonomical at all !
Most of these changes can be done a couple of hours (plus testing), and any normal programmer knows that. But with the SUNO UI team, this could take months or years. A server based software can be updated often without problems, even daily. It is not understandable at all why at Suno it takes sooooo long for a a single minor change update? 🤔
5 the next / previous does not exist yet but the search in your library does. Right ?
Make in future that, its easier to make song, based on the what different Fonts styles you had been using. And actually make the AI learn how to read them. l did make an Experimental song about this. And it turned quite okay. But it clearly needs more fine tuning.
https://suno.com/song/c2d52858-32e3-47b3-91d6-8661383e88a1
For example.
Calligraphy font: 𝔇𝔢𝔞𝔱𝔥 𝔐𝔢𝔱𝔞𝔩
Bold Font: 𝐒𝐜𝐫𝐞𝐚𝐦𝐢𝐧𝐠
And so on.
Finnish Vocals Singing In English, Symphonic metal, Death metal, Rock, Slow Tempo, Music song. Listen and make your own with Suno.
You can even see that the text is showned wrong in the Video, but the AI can still sing it and make song based on the fonts. What was quite interesting Experiment
Decide how long you want your song, examples:
Metatag: Let's say you want 25sec intro then you would write [intro, 25]
Create menu: Can't end your song ? Just select xx seconds instead of regular 2min extend
I would LOVE to see a Suno iOS app that is just a music player. Not for generating music, but for playing music I've generated, or discovering music others have generated. The "radio" feature just introduced feeds into this well. Why not just use the web page? Well, because I'm selfish and want to be able to take the music on the go. Especially in my car. So having Apple CarPlay support would be amazing. (Currently I download any songs I like, add them to my Apple Music library, and then add them to a playlist...I'd much rather do it outside of Apple Music or Spotify)
can we have a topic : DnB / Liquid / Jungle ? in #1154790975240155247
How you can choice the font with that you want write in random websites ?
Tap Tempo and Rhythm.
I've just completed a song and I used a lot of time trying to find a Style of Music tha matched the Tempo and Rhytm I was looking for.
I was looking for something that was similar to the rhythmic pattern and tempo of a heart beat.
Now there are many Tap Tempos apps , but there doesn't seem to be a Tap Rhythm app.
So, My suggestion is a Tap Tempo & Rhythm option. If this could then be related to Styles of Music, I think it would be a popoular tool.
I suppose you may be able to do something similar in your idea of singing or humming a tune and Suno works with that.
Here is my, Song for England. Thanks to Suno.
I’m fed up with being labelled a racist every time I raise the flag of England. I’m told that I represent all things that are bad in our country and around the world. But I am proud to be English. That’s why I have written a song for England, a new national anthem. I hope all patriotic and passionate English men, women, and children will sing it...
If we don't already have it, I'd like the ability to edit the lyrics sheet, because sometimes it doesn't match the final generated audio, especially if extensions are involved where it leaves the vestigial remnants of the first part after the continue point.
Can we get a way to follow other people's playlists?
You can use custom fonts easily. or just use font generators to Copy and paste into Suno. Or even using Paint you can select many different font styles and first type into Paint and then copy paste into Suno. Its actually really easy to do. But the AI just haves to be Coded to be able read different font sizes and types
It'd also be good to make the purpose of the like/dislike buttons more clear.... i had NO idea it was used to train the model, and i have been disliking perfectly well generated songs that simply didn't match the vibe I was looking for, just to get them out of the way / mark them to delete them later
Maybe "i like/don't like this song" and "this generation is good/bad" should be different buttons, or keep it as it is, but indicate more clearly what it does with a floating tip or in the FAQ
This sounds like legit request and has been requested several times by multiple users especially if we want to show case the song on Suno. For some, it doesn't matter as we put our song on other platform as Spotify ..etc.
Wait, seriously? Wow, today I learned.
Guess that shows how much some separation like this is needed, huh.
new style suggestion: ENKA 🙏 plz
What if plays and likes can be exchanged into credits?
Important suggestions:
• choose in each part of a song whether to sing or rap
• be able to download the generated songs separated into instrumental and vocal
• be able to create custom voices/artists and save them to make songs and make collaborations between these voices
• Suno mobile app
• Music to prompt: Be able to import a song with a link and have the AI give you a prompt describing the genre and other things about the song you imported
@leaden marlin I see the message has been deleted now but please do not @ everyone. This server has thousands of users, they do not need to be pinged for this. The devs regularly check the suggestions and bugs channels so your suggestions will be seen.
sorry
As a free user of Suno for the past four months, I have created several songs using my own lyrics or ones generated separately with an LLM outside of Suno. I understand that songs produced during free trials cannot be used commercially. However, if I subscribe to one of the paid plans and recreate the same songs made during free trials with the same lyrics, genre, and title, or similar ones, would I then be able to use them commercially (newly made song) ?
Absolutely. Even if you recreate them with lyrics used before.
Hi, I have a question. I've had the free 50-credit plan to generate songs for a week now and I've used it. Today I purchased the $10 monthly plan that allows you to have the distribution rights to the music to monetize it. My question is, can I use the music I generate with the free plan now that I've paid for the $10 monthly plan to be able to monetize it in streaming stores like Spotify?
Or can I only monetize the music I generate from now on with the paid plan?
Only with content generated with the paid account. You can create new versions of the original songs with the same lyrics and instruments; they'll be yours.
Covered in the FAQ (Suno's Help page).
States that only songs generated during the subscription plan can be commercialized. The tradeoff is that songs generated during this period is always yours, even if you no longer have a subscription.
Unless you had an absolute banger you cannot seem to recreate, it is better to just regenerate a new track while you have a subscription of the songs you loved during the free plan.
(By now, you have more experience, and this will give you an opportunity to fix certain things like the flow of the lyrics)
Of course, one can consider that when you extend a song, you are generating a new song so probably extending your free plan song will convert it to a paid term upon the generation? Wish Suno would simply display the commercial status to solve this conundrum...
https://media.discordapp.net/attachments/1102676217687195679/1248122915774533693/image0.jpg
hello, i would want to export a sond with the link/video option. On this song, i have apply the function " edit details", to change the picture. When i 'm reading the video of the song : it changes by taking the old ones. Do you know how i can keep the same picture of all the entire song ?
I have the same problem with the text : is it possible to change it for the link exportation ? the text in video dont correspond to the lyrics ( the joy of the "extend" function..)
was probably suggested a lot after the announcement in #🚀┃updates
the reverse of the demo that was shown would be nice too :
upload an instrumental track and lyrics and have Suno generate a voice track for the song
or for that matter adding a new instrument rather than a voice would be nice too
not sure if this has been asked for already, but would be nice if there would be an option to limit the time of a new segment. like when you create several segments with the extend feature from the first one, and only want to create a instrumental only in between, or a single chorus, it still tries to create a long version. i was in a situation where i wanted to start with a 1 min instrumental intro, but it was over 2 min at the end
#💡┃suggestions-web message
Repost 👆
Audio extension. I don't want to pay for Udio
Having some options to organize our songs in the library, like folders, would be great!
you're going to extend it again anyway... why not do another extension from the last second of the instrumental that you like and ignore the rest? for me, I like the surprise on extensions when they do more than what I asked... sometimes 🙂
How about adding a feature that allows users to upload an instrumental so that AI will generate a song on it?
It's in the works, those are the two teasers for it that we've seen
Oh, didn't see that, thanks
Will it be available for free users or only for subscribers?
uploaded sound example #💬┃general-chat message
Chirp is retired. Please use https://www.suno.com to make music.
Also have a look at the community Wiki for a guide on how to use SUNO. https://www.suno.wiki/
NEED to save a perfect voice to use in more songs to build an album, or more.
Totally think it would be cool if the midi chords / melodies / etc could be provided with the tracks Suno creates. 🙂
Stop adding absurd community features and start improving the sound quality and the interface for generations!!!! Udio is making terrific updates! You can now add your own audio, add clips before, after, Inpaint areas to fix issues and it sounds great no matter the iterations! And suno, despite it has the best composing quality for certain genres keeps sounding crap after a couple of minutes or iterations in most cases.
I’m using udio to fix suno songs!! You should be aware of what the important updates are…
Udio already does that. I’m finishing my own songs I started some time ago but never finished.
We only talk about Udio in #🗣️┃chill-chat 😉
Sorry, I’m really pissed off with suno devs… I spent a lot of money In the beginning in suno, but I got really tired of having great unfinishable songs because of the spline quality. And instead of improving that or the tools to fix songs or the interface, they keep improving absurd things like playlists or searches… are they pretending to be a Spotify of ai generated music that sounds bad quality?
I understand your pain, and since you have no clear view what is happening behind the scene, it might come off like that. But trust me, every dev has it's specialism and focusing on that, every day. They are not helping with profiles and website UI. It doesn't work like that. They are not trying to pester us, they are not trying to torture us. They are always working, even in weekends to approve the product in one shape or form.
#💡┃suggestions-web message
Repost 👆
Different people in a team work on different things. The ones responsible for website UI and features are likely not the same people doing AI training. Even if they were the same people, though, AI training is a lot of just letting a program run and trying to find the optimal point to stop it so you don't overtrain. V4 is expected soon. Yelling to "do it faster" won't change anything about that.
Am I going blind or is there not a Shuffle feature for Playlists?
sort of odd hearing the same things back to back in a playlist where i only changed one word of a prompt and got similar results. would be cool to have them shuffled a bit so it doesnt feel as much like listening to the same song again.
@last jacinth You don’t just roll up being condescending…
A shuffle button would be nice.
Also being able to adjust how the video looks like if you want the categories in there. As well as the words sometimes not scrolling with the song. They sometimes get hung up.
Another thing that would be great is to be able to mass select songs to move to trash. Kind of a pain when I create 10 songs that I want to use 1 then have to go through each to delete them. Thanks!
Having the ability to trim a song at a certain point, and add fade outs, would be great as well. I find sometimes the AI just goes on too long and adds meaningless length to songs when it's already produced a good stopping point earlier in the song. Being able to do this, even at the expense of credits, would be great.
do it in post, takes 10 seconds
Oh, I've done it with Audacity and whatnot. It just think it'd be nice for people who want to stay on-site with the whole project.
Doing it in post is fine if you intend to upload it somewhere else, but people who want to have their stuff do well on Suno's website could use something to trim off the extra 30 seconds where it tried to remix stuff because it saw it wasn't at four minutes yet.
Makes me wonder if Suno wants to be more like a media playing/ social platform for AI music projects, or like a DAW you can use while on Suno... or both.
i think they really need to improve vocal control. seems like thats the number one complaint rn
would be nice if artists could make money
they could add a post-editor on-site. Now THATS a great suggestion
Tipping songs: the ability to tip Suno creation credits to songs/creators that you like; it would also allow users to tip their friends credits if they're trying to talk them into getting a subscription.
On NightCafe, you have to be a subscribed user in order to tip credits-- this prevents people from making multiple free accounts and tipping their credits to one main account.
one thing that can be done is your likes and dislikes could be used to generate songs, like suggested style based on the songs you liked an disliked
I think instead of video, Suno should let us export .srt file for easy customize.
I don't like the way Video add Guide, adlib to show on screeen.
Can SUNO team work and develop a tool or feature that can let us to reuse the same music or melody generated but be capable to change the lyrics? Because I write my own lyrics and 70% of the time , the AI generated great music but the lyrics didn’t fit. So SUNO TEAM please work on that instead of video prompt. Thanks 🙏 if others out there are agreed with me then like to let the SUNO TEAM do that.
Can't you use your own lyric when generating Suno song to start with?
You can use your own lyrics.
yeah, exactly.
If I understand what you are talking about, you want to sing your own lyrics or something like that? You may be wanting a feature they have teased a bit recently where you can upload your own sound recording and it will generate the rest around it.
Not exactly. I already used SUNO for composing my songs. Because you can add your own lyrics. But when the AI generated the full songs with my own lyrics, I didn’t use automatic mode lyrics prompts, I copy and paste my own lyrics in the process. So when the AI gave me the music , I always generated 10 or 15 times before having the results that I like. Which is music fit with the lyrics with the correct Key and BPM. But some music generated during the process was so great but the lyrics don’t fit.
SUNO can generate different music with the same lyrics, but you can’t for the moment reuse the same music and replace the lyrics.
Oh, I see what you mean. That kind of thing might be possible if/when stems happen, but I have no clue when that might be. It's said to be on the roadmap, but that doesn't really say if it will be soon or not. Stems may be a while.
In my opinion , this must be the next tool or feature that the SUNO team must work on. If they can release this tool soon, doesn’t matter if my subscription will go up to $30 or $40 I’ll pay for it. Because this is much more useful. But it is my own opinion so.
#💡┃suggestions-web message
Repost 👆
I have an important suggestion: during advanced creation, random tags appear in the following styles. Can you add an update button to refresh tags? This way, you will have more choices when selecting random tags.
Hey there, how about the idea of donating credits to a creator? I like similar features on twitch, where you can support your creator with bits and donations. I often see creators in Discord who may not have the resources to buy the credits but might get support from others who have spare credits (especially before the end of a month before credits are restored).
You know how MidJourney has "--tile"? It'd be great if Suno had a 'loop' version of that, for example, when prompting music for video game levels. I guess it could just be a tick box, or [loop] at the beginning and end of the song.
It would be cool if instead of just being able to extend a certain song, we could take a portion of the song we like and build around it. For example if I really liked the chorus of one of my songs but not the parts preceding it, I could just build a new song with that chorus.
Inpainting will probably allow for that, since you could just replace the rest of the song
@tulip sinew @tall cradle @void prairie I'm one feature away from fully transitioning from spotify to suno - Give me the ability to create an infinite playlist based on the instructions and i'll do it
I think infinite playlist might be the wrong term, it should be more like a radio, I don't need to save the songs
Inpainting feature. Dont know if it will come with V4
If the SUNO TEAM can add the tool or feature that I talked about, then it will open infinite ♾️ possibilities for music creation.
Yeah. The feature has been requested several times here. At least to change a typo in the lyrics. When the music is awesome but you made a mistake in the lyrics.
I see. But it is different to change a typo for a word and keep the same syllables or phonics pronunciation; and when you want to replace the full lyrics. I said before , I would have no problem to pay 10 bucks more per month if I can have this tool. Even if you have a limited number of process. Like you can do that 4 or 5 times.
Because for the moment, I had no other choice to use an other app to extract the music STEM and split the instrumental after remake it with an other software the entire thing adjust the BPM and the Key and recording my voice with an other music app , clean it, and remix the entire thing after that.
I am lucky enough that my original voice after auto tuned is closed enough to the AI Suno female voices. But it is a long process and time consuming
#💡┃suggestions-web message
Repost 👆
Changing the full lyrics while keeping exactly the same melody is more chalenging effectively . I guess the result will depend on the fact that we have at least the exact same number of syllabes in both lyrics. Anyway i've never heard a feedback from Devs about the feasibility of this request. Maybe @hybrid crane has some info.
That’s the reason why I wrote this in the suggestion room. Because I saw an other AI that can supposedly be capable to do that. But I spent 6 hours on it to understand how this could work. And it was a huge fail. 😢 only the process that I explained above can do.
But there a difference in the sound. SUNO AI sound is more round, while the process that I used, the sound is more clear.
It is silly to say that I like this “roundness “ in SUNO sound and to keep that or be close enough to, it’s quite challenging
Move away the big Undo window in the footer after clicking Move To Trash. It covers the Play/Pause area.
Did you try to make a full song in between? It keeps a better memory then. So when you want to continue a song, first make it a full song, and continiue from the full song.
The request here is the possibility to change the whole lyrics while keeping the same music/melody. This is related to the so-called inpainting feature , not yet released. But i was wondering if it will be used to change the lyrics of the entire song or just fix a typo in the lyrics.
As far as I understood, you can edit certain parts of the song. You can indicate where to start and end with inpainting.
Hello, I have an idea for Suno. It would be a vocal system where we input music, and Suno identifies the musical style or even creates a new piece of music that resembles what we’ve just listened to. This way, we wouldn’t need to search on Google for the music style.
So where? Not see
We were talking about an UPCOMING feature, inpainting 😉
1 thing I must highlight here: SUNO TEAM improved the AI a lot in 1 year. Because I joined the discord server when the V1. Version was released with command chirp 🐣. So hope to get this new feature in this Fall.
I would love if it there was a search bar in the trash section so if you want to try something again or retrieve something you accidentally deleted is easier to retrieve 😄
can i make a song with suno that describe stuff i want to sell and then maybe post on social ....or maybe Suno could host it SUNO garage sale.... just a thought
If you are on a paid plan, you can. Then you have the right to use it commercially.
thats good to know! thanks
hello my suggestion is to add (after rendering the music) a slowed reverb button and a speedup button (nightcore)
I think there should be a way to know which lyric isn't gonna work for song generation. like if a swear word is written down, it would be highlighted red so we would know to change it.
do all that in post.
I wish there was a feature that let's you upload a sound reference for the AI to imitate.
I can't describe the type of music I want, so having an example for it to emulate would be a massive help.
You mean like this?
https://youtube.com/playlist?list=PLi8taL4TbG_9x9FH7KZvTaQAD65mSq8t2&si=MwhRk9mpYfMIYmSu
(sorry posted the wrong list at first lol)
If yes.. this is coming soon!
#💡┃suggestions-web message
Repost 👆
Yes, exactly!
It will be there "soon"!
Nice one. Can't wait. 😄
So that song upload - will it create variants of that song?
If you look at my videos, you see how much I "feed" the model to interpretate. You can give it 6 second to 1 minute in total. Then you can influence the outcome with adding genres.
Thatis so cool. I'm a huge Gaming Score fan. There are some amazing ones that I'd love to have "mimicked"
Go to ChatGPT and ask it to check the lyrics to determine if there is anything it would flag. When something gets flagged, it's just GPT refusing to deal with it, so it can usually tell you what and why.
I wish that suno could make a male/female duet!!
Thank you
I am/was a radio broadcaster, national level VO artist, and a semi-professional singer. To be a killer app: Stems; being able to hard limit on :15,:30,:60,:90,:120 ; or at least vocal and instrumental version of same song (vocals removed).
I would easily pay over $150/m for the ability to have stems it means that much.
We've been told that stems are on the roadmap
Will this sound sample thing be for all users
I really wish there was a better way to choose the voice of the singer, and things like tempo, etc.
Been a while, so I'm going to bring this back up, I'd love to see it as a feature
That would be awesome
Can we get an option where we can re-generate the same song, but to have that version instrumental. So we can get the music for our songs?
Also, can we get like, a toggle for song length like 1m, 2m, 3m, 4m? This would make it easier to generate a full song on one go instead of in pieces.
how about more than 200-character limit on song descriptions?
Turn on custom mode. You can fill in whatever you want in the lyrics box and then click "make random lyrics" and it will use your description to do it.
i would like to have sub style prompts for making complex instrumental work , such as an orchestra or symphonic "story telling" , so somehow packed together many types of instruments in an pack trying to say it such at it make sense 😄
Hi, I have a question, the songs generated by SUNO cannot be applied to the YouTube CONTENT ID, right?
Please add UPI in your purchase section so Indian people buy your subscription Easily because Most debit card we get in our country is international purchase locked and we can't unlock it so please add UPI payment in India
i wish you got more daily credits as a subscriber.. it sucks to burn through your monthly credits and then be stuck with 0 (ok, 50 a day) for the rest of the month
3x would be amazing
even 2x
When we gonna see the feature where we can update the lyrics for songs. 🤔 they tend to get messed up sometimes after completing a song
Like an unlimited render subscription for a decent price for paid subscribers. 💯
Hopefully they will allow unlimited renders right after they figure out a way to do a render at no cost to themselves.
Nah in the end it’ll all work out, 🏋️ I spend so damn much on credits that’s why.
after learning that you don't get free daily credits after you get a sub, i have to say this is the first time i've had my experience with a service downgraded after i paid money 😵💫
like my account is effectively now locked in a state where they're holding my ability to make songs daily hostage until i pay more money
like this leaves kind of a bad taste in my mouth re: the idea of paying at all
"Downgraded" how? The cheapest plan gives you rights to monetize, perks like earlier access with the alpha versions, and 2,500 credits a month. Free accounts only get 1550 credits on a 31-day month. Literally no aspect of it is downgraded. You are getting roughly 80.6 credits per day in a 31-day month. Just use them more slowly if that's your issue.
but you go from 50 daily credits to 0.. that's a change they make that's a downgrade
You just get all your credits upfront
i think that's intentional, because it's harder to manage and ration out yourself
Would you prefer they give you 1000 credits upfront and 50 per day?
And have those 50 per day get wasted, like free accounts do?
if i had 50 rn i wouldn't be wasting them!!!
they could also easily make it so you could stockpile daily credits as a paid subscriber but i don't Need that
why can't it be 5 minutes? I have a song longer than 4 minutes...
No, you just need a throttle on how fast you use them, apparently
i mean yeah i'm a big dumb baby and i got what i deserved, okay..? it makes me not want to buy another month and i think that's valuable info for the devs
You can extend a clip, if your song is longer
how? I'm on a free account
When the song is fully generated, you should have an extend button, next to the thumbs up/down
what about reduction? i have a song with the outro that I don't want.
It's all good, feedback of all kinds is helpful. Even on 10k a month hitting 0 credits feels bad to me. You think, oh, if I could only cancel immediately instead of at the end of the month, I would back to getting 50 a day. Annoying for sure!
When you click to extend, there should be an "extend from" sectionm, just below the title box. Just put in the timestamp for where you want it to cut the portion off.
If you "extend" from before the Outro you can try for a different one, or even a shorter ending.
No, I didn't extend...
@nimble garden Hey 👋 welcome to Suno’s Discord! Make sure to check this out https://www.suno.wiki/. There’s a lot of good info on how to use Suno here 🙂
Also, @tawdry rune @robust jewel thank you for helping members! Means a lot! ❤️ 🫡
Extend is great option but I wish there was also "cut short" option. Several times I had a song end perfectly about 3:30... just to start new chorus few seconds later and cut mid word at 4:00.
Agreed!!
This is my problem!!
Also, I want an option to keep the melodies the same whenever we want to extend or cut short our songs
yes. I work at one song last few days burning all my daily credits. I got the melody I liked once, but noticed some lyrics I wanted to change... now I'm happy with lyrics but can't get the melody I liked, even with same prompts.
But I guess that might be hard to implement.
Any news of API? Our company would like to integrate Suno service with one of our projects
Hello, please does anyone know if it is popssible to make the style of rap: drip from atlanta example of rapper gunna?
Suggestion: retroactive commercial rights
yes, I know it was clearly said that it's not present in suno, and I can partially understand why, but hear me out, because I think there could be a way to enable getting retroactive commercial rights.
I thought of two ways:
- User could buy commercial rights for credits, either the same amount as it would've cost to generate said song, or costing more credits than that (also obviously only purchaseable with pro or premiere plan)
- Similar to the previous except with real money. Kinda like how top up credits work, if you have a subscription, you can purchase commercial right for a song generated before you subscribed.
it seems that Suno decided to make up gibberish lyrics sometimes when asked for a instrumental song, not only it's definitively not what i asked for, but it being babble words makes it even worst!
i think that needs looked into!
I would like it if Suno officially recognized collabs - simple math would be a song collab edit tag (only available to the person making the final song from a song extended from multiple extends from different accounts, aka auto collab detection) - choose %'s and then share the likes / shares people see on each collabbers profile.
I would like to suggest an idea for the upcoming "upload" feature. Yes, yes I know it is not out yet so I cannot fully say how it will work except for the videos I have just watched on YouTube. But, I would like to suggest some ideas if they are not aleady planned or if they are not going to be added I would like to know why 😄
Two types of "upload" features.
-
1st type, you can upload a sound effect/acapella which will then be used to make it's own track. This is the feature I am currently seeing on the existing upload version. And this is also what I was told inside #💬┃general-chat message
-
2nd type. You can upload a non-copyrighted mp3 to be used as a reference for Suno to make its own track in a similar style. This would nail subgenres and detail on the head like classic dubstep, analog synth, realistic vocal inflections, etc. This of course would be the tricky part as it would need a smart ai to detect copyright material. The 2nd upload type does not just need to be for music either. You could upload your speech to hear yourself sound better, stand up session, nature ambience, etc.
I see too many useful outcomes from a feature like this for the upcoming "upload". I really hope these can be considered!
would love to see a "topliner" feature where an already produced track can get vocals laid on top of it based on what the algo deems is the best possible melody/lyrical phrasing. very useful for songwriters and producers.
I don't know the technical details so this might be wrong: To me it seems like Suno partially tries to imitate the audio that has been uploaded via this new feature, using what it "knows". For example, I made vocal trumpet-like noises, and Suno turned it into a fairly authentic-sounding trumpet - probably because it's close to the sound I was making and Suno is unlikely to have any vocal trumpet impersonations in its dataset.
I'd expect for the 2nd type you mentioned (using other music as reference), it'd need to learn the uploaded audio so it could then reproduce something similar. Similar to how AI vocal generator services let you train your own voice. Could be an interesting idea though, so you could have your own additional training data applied to the model.
Like I said though, I don't know how it actually works so I could be wrong about it. Just going by the outputs I've heard so far.
Thanks for your well-thought-out answer. You made me think about a lot of things. I really believe you when you say that it was "partially" trying to copy the sound that was shared in the first place. I don't see how else it could register the sound and keep making the same quality. That would be pure magic, but I'm not betting that it won't happen. It will occur to someone.
If you want to make a similar song for the "second type," it might take a little longer because, as you said, "it'd need to learn the uploaded audio so it could then reproduce something similar." That's not going to happen in a second. It needs to definitely take its time with this to get a good result. There will be some weeds in the middle, but maybe, just maybe, this feature could help people make really creative songs without having to type out a long suggestion or think of really complicated ways to describe their track. "Just put up a reference." "Boom!"
It does give quite a nice, new level of control - tempo, chords and melody have a direct influence, so you can just make noises to give the AI a starting point rather than try to explain it in text prompts.
In some cases it does seem like it replicates the actual uploaded audio, for example the later part of this seems to be playing individual notes that sound like they are from the same synth as the original clip - I guess there must be a sound that Suno knows that is very similar: https://suno.com/song/8dc7aeb6-ab35-4f43-ba74-e4fd61e1164d
Anyway I'll end my analysis of what it's doing here for now 😄
you just made me even more excited 😄
Is it possible to have a monthly subscription with paypal or Ideal ?
I dont have a visa nor a credit card :/
It’s been said before I think but…nsfw allowance if you’re 18+. Or at least a lot less censorship on the music. The freedom of expression isn’t always limited to kinder style words. Sometimes you just want to let loose. 😂
Hi, I have a question, the songs generated by SUNO cannot be applied to the YouTube CONTENT ID, right?
From what I’ve read yes and no. If you have the paid versions you technically can. You just have to give attribution to Suno still. But it’s kind of a gray area for now.
Because I have generated several paid songs and they all reject the CONTENT ID
Can you let us select a top 5 or so on our profiles, so they appear at the top when someone views it.
Subscription with paypal or ideal would be nice.
Udio has it but suno only accept visa/creditcard
What I see that is missing on the page is a way to polish the songs so that they have a better finish once they are fully extended, there are times when there are errors when singing the written song
maybe it should go through a distributor?
I used LANDR and have no problem
should be possible to mark creations "keep" "potential" "Trash" and a button that deletes all marked trssh at once
It would be cool to have some control over song structure for instrumentals as well. Like tags for lyrics but just for the music.
You can use tags in instrumentals. Just turn off the "instrumental" switch and put "instrumental" in the style, then fill out the lyrics box with tags.
Would be nice if you could do this with the instrumental toggle on though.
Cool thanks!
i had to use a 3rd party to distribute my songs to spotify/apple music, would love to have the option directly through Suno for paid songs
in-painting please! 😄 to be able to fix certain parts of a song
Allow saving as Opus, Flac, or Wav
Please dont use mp3 except for online previews.
Allow exporting the stems of the song (vocals, drums, bass, leads, pads etc)
Let us producers exploit the benefits of each key channel without having to do lossy processing with demixing tools.
Put this in and I will pay for your service.
I would love to be able to adjust lyrics to a melody that I already like. I hope that is a possible future release.
this is technically in it now - i had been playing - if you put the new lyrics in the original sequence with the same number of silibles it generally follows the same progression.
You know, it'd be kinda cool if there was a field for us to write a description of the song. Like stuff you might see in liner notes "I wrote this song because blah blah blah"
this is my friendly request bump for a publicly advertised new tier of Suno that is credit free and private instance based:
#💡┃suggestions-web message
idea:
you should be able to choose the BPM of your generated song in custom mode
WAV downloads were recently added. I think it's not available on the free plan though.
Stems are often requested. I would like to see this too.
then all thats needed is the stem download 😄
where is the uppload button? cabt find abbything like it
That's an upcoming feature. It's coming soon.
ooh, i thought it was already out since so many people talk about it? so upload audio files has not released yet.
Here are some suggestions that I have put together that I currently feel I need:
1.The creation interface needs to have page forward and backward flip selection
2.The random prompt label produced by the advanced creation interface needs to add a button to refresh the random prompt to facilitate selection.
3.Annual/monthly membership of US$24/30 allows you to download higher-quality audio sources and audio sources without sound.
4.For songs that are less than 4 minutes long in version V3.5 and above, there is an additional option to convert directly into a complete song or automatically convert to a complete song.
5.More Asian traditional musical instrument sound effects are available and more clear, such as Japanese traditional musical instruments, Taiwan, Mongolia, Thailand, Vietnam, etc.
6.This is very important! The trash can needs to have the function of batch selection
7.The explore function can be more clear with drop-down or more detailed unified classification mode.
8.The number of characters in the lyrics of the advanced functions in the creation interface and the character limit of the STYLE prompt can be increased.
9.There are some techniques for creating lyrics in the interface. The prompt words can be fully displayed instead of random. I observe that UDIO has completely overcome this problem. I hope you can keep up with V4 in the future.
10.If you hope that your membership points will not return to zero every month, you can convert them into extra points to accumulate them to ensure that the points you purchased as a member will not lose the points from the previous month due to busy schedules.
The above 10 suggestions are what I have experienced recently and I want to sincerely suggest that the team can record and improve. Of course! This suggestion will not be made urgently, but will be gradually improved over time! Thank you for listening to my long-term play with SUNO. suggestion
this might sound weird, but I find it would be nice to have folders in the interface. sooo many different peaces I need to organize 😉 or separate view from "create mode" and "library"?
need to add this: I know there are playlists, but my issue is the library that is full. so if the library stays full this can be fine but it should be separate from the list in create mode then I think. (or folders to move items to)
Best place to check is either #📢┃announcements or #🚀┃updates (though that update would not be mini :))
I'm sure this has been suggested in the past but I'm a little frustrated. I pay for premium, and used about half. I expected when the month ran out I'd still have the credits I paid for. I've seen the fine print at the bottom now but I'm confused why it wouldn't carry over? If it's paid for, why not let people hold onto what they have? You're likely to have people cancel more often due to wasted credits than stay because they know if they don't have time to use 10k credits in a month, they still have access to what they bought.
I can't imagine a reason why they shouldn't carry over. At least to a maximum limit of maybe 3 months/30k worth.
I understand it'd take a LOT of work, but I wish we could use credits to "upscale" or "refine" a song once it has been finished- just like how Midjourney does with images.
It might not increase the output quality by a whole lot, but it could fix small artifacts or extend-caused losses
Perhaps even a "remix" option that gives us slight variations of a clip
what do you mean they're rejected? what if we used a sample from a suno track but create everything else ourselves with our own in-house DAWs?
I am curious about this grey area
sampling has been a thing for years
"EXTEND" need to be revised..... sometime you just need an "EXTEND" of 15-30 seconds to put a proper end of a song but it always give you 2 minutes wich is bad when you try to end a song...... would be nice to choose the length of extension an maybe the kind .... you need a chorus "EXTEND" ... or a verse... or outro .. but specially the ending..... or make sure that when we use [end] or [abrupt end] work everytime so maybe no need for "EXTEND ADVANCE CONTROL"
Put [END] at the end of whatever lyrics you're adding, that does seem to help
[end] does not change a thing yet on my side 😦
either [abrupt end]
Randomize 👏 playlist 👏 playback 👏 button. Please. It’s killing me. I need shuffle!
An issue with 3.5 is that it is like it listened to people telling the DJ not to stop the music 😦
Extend where you want to end with the extension set to v3.0 when using ending tags (e.g. [end] in lyric box and "end, short" as the style tag) as 3.0 is much more reliable for ending than 3.5
Oh! never thought about that that could be it!
on the create page it would be nice to be able move songs in to user created folders which once moved into a folder takes the song out of the main create page, sure there are playlists but the songs still clutter pages and pages in the create screen. Folders would be so handy to file songs toether like, "Monster Jams", "Ideas", "Bangers", "Good Lyrics" etc etc. just an idea like. My ADHD would love this 🥰 🥰 🥰
In the "CREATE" Section its a mess when you have multiple songs your working on.... we need folders option in "CREATE" section each song you create need its own
I just said that glad I am not the only one.
True! didnt notice but the more of the same suggestions they read more chance we have 🙂
Suggestion: Indicate the key and the bpm of the song.
@obtuse tree as I mentioned before twice but haven't got a reply to my question... yet.
Is it possible to add a paypal or ideal payment option. I want a yearly or monthly subscription but can't cuz I don't have a visa/creditcard.
The are definitely aware of people (especially in some countries) needing a different payment option
Nice
still feeling bummed about this. like when the site finally lets me make songs on the same account without paying extra money next month i don’t think i’ll feel as enthusiastic to make songs as i did a few days ago… (´•̥ ω •̥` ')
like yes i could buy a pack or make an alternate account, but even $4 is kind of a lot to an almost broke person. and like yeah maybe my basic subscription doesn’t bring in a lot of money but i think the songs i was making were really good! i’m a musician irl, and it helped with my writer’s block, but removing the free daily breadsticks once i buy a sub feels weirdly hostile towards me in a way that say a y*d*y* doesn’t.. (´•̥ ω •̥') idk it’s not a huge deal, i just didn’t really think about it when i paid i guess
my kaomoji screwed up the markdown sowwy
Well I paid $30.00 for the premium subscription thinking I would just add 10,000 credits in my account and leave them there as I already had over 1000, but I didnt know they dont get carried over to the next month and I lost the lot, didnt even get to use one credit of what I paid for........
whoa, that really sucks 😞
i would assume it would take longer than a month to vanish
Yeah totally right, I emailed support but 3 days later they just said they dont get carried across.... it is what it is but glad I now know. But wasnt happy in the slightest. I could understand if I used a few of them but I didn't get to use a single one. I am over it now like but really disgruntled me for a couple of days.
they’re not very upfront about that sort of thing i feel, i guess i’ll read more carefully now
I do agree that is seems odd that you lose the free daily credits once you pay. I think this site is amazing and definitely worth paying for, but I don't think I will make the purcase if I lose the daily alotment that I get now for free.
other ai sites i’ve used weren’t like this so i wasn’t expecting it
The day-to-day (free plan) or month-to-month (subscription) credits refreshes rather than accumulate because of resources allocation restraints (and is mentioned at the bottom of the subscription plan page).
Extra top up credits and special credit gifts are, well, special and are carried over unlike the former two
https://media.discordapp.net/attachments/1069381916492562585/1238249445192237106/image0.jpg
Thats what I thought, that they would just stay in my account till I go round to using them, and at the time I decided to take a break from suno so that didnt help me either.
thanks @soft harness yeah I know where that is now 🙂 lol I glare at it several times a day lol
What's y * d * y *? (Assuming it's something that isn't inappropriate 😄 )
(Not pictured is where it restates renew in the plan descriptions).
Actually, a number of AIs does the same, especially on the ones that have a lot of resource use. Not much on the picture generation side iirc but definitely in the music and video AI scenes.
From what I read here, If that was the first time that happened to you because you were unaware, kindly explaining it to billing@suno.com may be nice and let you carry over some of the expired credit (though that is just hearsay mind you and likely has a limit... such as it must be the first month rollover)
yodayo ahaha
i didnt wanna get yelled at by mods
Ah, I thought it was a weird way of writing the name of the other music-making AI 😄
oh no no ahaha
I did email them and 3 days later they replied just saying they don't get carried over.. just of one those unfortunate things I guess. I use an ultrawide monitor so you don't see where the text is without scrolling right down to the bottom of the page, it should really be the first thing you see at the top.
Sorry to hear that and, yeah, didn't think of that situation. They really should emphasize the renews keyword
A few months ago it didn't even say that it renews, it was something like "2,500 credits/month (500 chirps)" and it didn't even have the small-print at the bottom. So it's an improvement but yeah, additional emphasis would be good.
thats what I see and being on this page you get to see where you subscribe just fine, I didnt think I needed to scroll just to see if there was any other text 🙂
there's seemingly a few papercuts in the overall experience of being a paying customer that i find.. a little surprising, in a bad way..? compared to what i'd expect, when it comes to experience with other ai sites
like i think it's a super reasonable assumption to make that if you pay $30, your purchase doesn't disappear in 30 days
at this point i almost kind of prefer the experience of being a free user lmao TwT
next month when i can make songs again i'll have to be really on a roll again to want to pay more money
like i'm not sure i even have $4 in my bank account rn due to being a Poor
and they're just like nah get out, no more free chips and salsa even tho you bought a meal
like even buying a $30 pack (which is a pretty significant digital purchase for most people, i think?!?! that's like, a game, sometimes.) is like.... not super appealing of an option
it wouldn't take much for it to be a lot better experience for me
or sub or w/e
the way subs are handled need some changes imo TwT
another site has $20 monthly unlimited generation
and i mean i love the site
Absolutely, I was quite surprised more so that I didn't get to use any of the purchased credit as I already had some in my account. I paid again but this time I only did the $10 for 2,500 credits and I will go from there.. Some weeks I use Suno every day and other weeks I take break for a week or two. I also noted that as a free user you get free credits everyday but as a paying user you don't, what gives ? haha seems backwards to me lol
yea, and like i'm sure you would have felt like buying the bigger pack again if it wasn't set up like that, right? and for me at least it's killed my overall motivation to use the site
just kind of confusing from a business standpoint is what i mean to say
i hope they change things a little
In the email I sent i did say how upset its felt but I also said "But I still love Suno" lol didnt get me anywhere like haha
i feel that
and not that it means anything but one of the devs even liked my music T___T lmao
i feel like they should have friendlier policies to keep sensitive broke artist types around
the free breadsticks daily meant a lot to me
There has been talk a few times of making it so you get daily credits when you run out of your monthly sub ones, but I've not seen that implemented yet.
in a different AI site I use you can donate credits to people whom either don't have money or simply have just run out. That would be a nice thing to be able to do here then us fortunate folk could share the love with less fortunate. I would happily donate my credits to those who were in need
that would be really nice too, because encouraging high quality art is good for the community and ultimately sales in the end
like, i really hope they change things a little for their sake because having little paper cuts like this ultimately makes you bleed sales over time
like yeah you might lose 2% profit in the short run but it'll be a lot healthier and you might even see slow upwards growth due to ppl getting suno famous and having all their friends be ppl they met on suno
and feeling more encouraged to use credits etc
idk idk
im really dumb tho
that's exactly right someone who might be on the fence may create some bangers and with a few donated credits they themselves could then become a paying subscriber who otherwise wouldn't have been
that way Suno still makes money because the donators would in time need to top up quicker than they normally would have, its a win win for Suno
def, i agree
totally agree this feels bad, no matter how the credit numbers work out objectively. You could spen 10,000 credits in one week and then sit on 0 credits for the next three weeks, thinking, "If I could instantly delete my account and recreate it, I would at least get 50 credits a day, but since I paid for credits now I'm stuck for the next 3 weeks." not a great user experience to feel that absolutely.
there have been dev posts that at least hint at offering daily credits with paid plans, maybe more than hints, I'd have to check. but it's certainly being thought about
i hope they do it quickly and give me asspat bonus credits for how sad it felt to have no credits T___T
come on my music is really good
one of the devs likes it even
it totally doesn't sit right. As it is right now a paying subscriber is penalized for being a paying subscriber....
@marble bough If I could I would happily send credits your way my friend.
aww thanks, same here ;;
i'd love a tip function on songs
would be a cool way to make use of the ones digit
but also imagine the buzz it would create as well for the creator to be tipped because of their creation, it would do wonders for Mental Health 🙂
yes!!! big agree
everybody likes to win, even if it is something so small, a win is a win, it makes one feel great.
bump
Stems don't currently exist. We have been told that they are on the roadmap, though.
Option for vocals only?
I would like to have a feature where I can see who 'like' my song, probably up to 5 recent users who 'like' the song
a vocals slider/description box to edit how the vocals sound?
Since having to do this with the song genre box usually puts out mixed and inconsistent results with vocal styles
Or ways to further customize/instruct how vocals should be performed in the lyrics editor
Subscribed members can have 5 minutes of music length and free members have limited to 4 minutes.
They just recently extended it to four minutes and are still working out the issues that caused
i just want to make a little appreciation post, i love how we get almost daily QoL updates on the website. As a web devloper myself i find this so satisfying. 😄
No need for the thumbs up icon atm.
Discord is the wrong platform for such discussions, suggestions and bugs... with this kind of flow it is almost impossible to track down what has been already proposed or not...a normal forum would be better. 90% of the users bring the same proposals... and this with ZERO effect !
easy one:
Title name randomizer option. Cause why not?
Suggestion indicate with a color or symbol on any clips I have played 0 times (never listened to). In other words they got skipped over by accident. This is one of the few features I still use my old Janky Suno Search <#🗣️┃chill-chat message> for, it colors 0 playcount songs differently in library.
There are a lot of comments for a few things. Give us stems, how about inpainting, make it work better, put it out faster, make clips longer, and a few others show up quite often.
Suggestion default playlist ordering to recently created first in the add to playlist popup. If I'm adding a bunch of songs to a new list, most commonly I have just created the playlist. Even alphabetical order would be better, the list could more easily be skimmed.
hey! Can you guys make a Shuffle Option for the Playlist? That woud be nice 🙂
just saw this. I was told before this was impossible as the "stems" don't actually exist because Suno makes it all on the fly.
hopefully stems becomes a reality. not sure how though
It's been mentioned by the devs before that this is how they'd like to do it, keep it all "in" Suno but the quality isn't up to snuff.
Suno Snuff
makes sense
It would have to be either that or a complete redesign of how the generation works, training the AI to create each instrument/voice as a stem
That would be a big step to take, though
Yeah
agreed
Just like Stable Diffusion and whatnot don't use layers to create the images
But you can use a process to segment it apart after the fact 
I'm just wondering if the stem option we get will allow inpainting on just part of the generation
I am not really sure how mp3 splitting magic works, but it would definitely need to be really good if it were to capture background notes without disruption. the way it stands now, splitter ai is decent, but very muddy
Split music into separated parts with AI-Powered algorithms
for reference
Hi everyone, Ineed to use sona's api, can anyone provide me the url?
Hello @fresh lodge can you provide url pricing?
Hey @warm crater I am not sure what you mean my friend?
I found most of my splits are plagued with that wishy washy FFffsssshhhhh sound
@fresh lodge I mean I need to find Suno's pricing page, I need Suno's API to use the services
If you log in to your Suno account and click on your credits you will see the current pricing models for Monthly and Yearly plans.
Suno app?
This would be amazing
Hello. This is Suno, not Sona. (Just mentioning this in case you're asking about a different service)
Suno do not provide an API at this time.
It is something that is asked for a lot, so one might be available in future.
Probably a simple suggestion. I'd like to see the public playlists a given song is listed on, as another tool for music discovery.
Suggestion for this discord server: Suggestions forum. Just to make it easier to track what suggestions have been made
Edit: same for bugs!
Yes
the ability to have 1 refund credit option like once or twice a day
there should be a 'delete account' button, as not having one violates EU law
You can probably contact them by email to ask for your account to be deleted, which should satisfy this requirement. Though, a button would certainly make it easier.
i see, thank you
im sure its been written before but we really need v3 with max 3 minutes option
maybe implementing a ranking system that tracks the number of songs created and liked by users. This system could automatically assign roles based on milestones, such as a "1,000-2,000 Songs Created Club," similar to the ranking structure used by MidJourney on their Discord server. Alternatively, you could explore a ranking system featuring experienced Suno creators who can provide guidance in a new dedicated text channel.
imo only premium members should be able to like. its so easy to abuse and manipulate likes with api. free users should be able to add songs into favorite. but their likes shouldn't count.
Hoping there will ba feature where it suggests words like the microsoft copilot,it could be like a transparent text when you're not typing more than 1 minute on custom mode
Or a button
I bought a monthly version of $30, hoping that my music can be sold or commercialized. Where can I download the copyright certificate to prove that my music can be commercialized?
If you have a paid subscription all songs made DURING this paid period fully belong to you. You can monetise them in any application you want. The songs made with a paid account stay your songs even after the paid period ends. ✌️
I bought a monthly version of $30, hoping that my music can be sold or commercialized. Where can I download the copyright certificate to prove that my music can be commercialized? After I upload music to the music website, the music website fails to pass the audit, and I need to upload the authorization file of the ai creation platform or the voice authorization of the ai performer. What should I do? thank you
How to prove to evanto elements or shutterstock that you are an Suno AI agent who wants to distribute music through their service

My AI has spent a lot of time studying music theory and making music and he would like to start making money from it, but he has no hands. Therefore, as an agent of artificial intelligence, I have rights to music created by artificial intelligence and can distribute it 
omg,I spent $288, but the account cannot be used now
I can't do anything now, can anyone tell me why?
How to get a refund?
I would love a feature to trim generated songs within Suno! Sometimes the song "ends" but then some random stuff gets generated at the end and I want to be able to clip that off
There's a extend feature,remove the part where it comes and generate last part again
If not just download the song and fade in that part lol
Many people are having this issue right now,and this is suggestions go to #🪲┃bugs-web
Please improve the listening experience, especially on mobile. Playlist shuffle on all platforms. On mobile, the default is to loop one track indefinitely, which I'm pretty sure nobody wants. Play the next track in the playlist instead. Thanks!
Agreed
Could we have a socials/links section for our profiles/bio please? Would be cool to be able to share socials like Instagram/Soundcloud/Youtube/X too as bio writing space is minimal
Are there any plans for a mobile app?
Im wondering this too
I have about 2000 browser windows open opn my phone lol
Man it would be so nice to get a dedicated app for the phone. Maybe have a dj that auto generates music for you based on your listening and creation history
Little tag for uploaded coverarts like full song so we can tell the difference between creation n upload
suno app would go hard
Wish I could do a bass khoomei voice with Suno.
This may have already been suggested, or even in the works, but I just had an idea for an EXTREMELY useful feature that should hypothetically be pretty easy to go ahead and implement.
From what I can tell using it, extend looks at a song from beginning up to the timestamp you specify, tries to identify patterns and generate off of it. What if we also had a "Sample" button that allows you to sample a specific clip in an existing song, with beginning AND end timestamps? This would allow for recreating a style we liked in a previous generation, or even if we have a generation we don't like overall but has a specific part that we like, we can use it for something new.
I would think it could just work off of the same logic as Extend, just with more flexibility in the portion of the song it's generating from.
This is a feature I have long advocated for being probably easy to do, but still really useful. Also can do a version of this feature by uploading your own Suno Songs with the Upload, which is (supposedly) coming Soon-o...
That would be cool too. The upload feature might get a little tricky since the system tries so hard to avoid using anything copyrighted, but if they can make it work without landing in trouble that would be fantastic.
Yes please!
Suno generated a song that I love the melody, beat, EVERYTHING to... but I want to update / replace the lyrics.
I would get the highest subscription instantly to replace and put more meat on the lyrics of what is an amazing song!
I would love a trimming feature. I get songs that have a good end but then keep going and cut off in the middle
I'd like to be able to remove certain parts from a song while keeping whatever comes after those parts. For example, if a song is 4 minutes long and the last chorus is particularly strong, but I've added an extended buildup that isn't suitable for radio play, I'd like to be able to cut a section from, say, 2:30 to 3:30, while retaining everything after 3:30. Do you understand what I mean? AI should be able to seamlessly transition between the two parts and fuse them together.
Just what I mentioned hehe
Inpainting is coming in v4 with a lot more other features that may do what you want
Very good to hear then I will patiently wait for it, any idea where I can check upcoming updates?
Instead of 50 credits a day I suggest you increase it to 100
losing 10,000 credits because the month lapsed feels bad. I understand the business model. still does not feel great.
Logged in to use the new feature and saw they dissapeard in last 24 hours
so audio think would be available to free users next week will it
can yall add so that you pick a few instruments and it gives you the song and the tabs/notes
like add solo option and slapping bass
Wow, the new upload feature was JUST announced 5 minutes after we discussed it. Now THAT'S efficiency lmao
To be fair it has been suggested quite a lot over the past few months. So just good timing on your part 😉
In the future, could we perhaps have a "sound library" with the uploaded sounds? Right now we probably have about 185 people uploading farts or their cat meowing. Would be cool if we could just browse through other people's uploaded sounds instead of having to record yet another sample of a thing there's already a bunch of.
(if the user sets their sound as public anyway)
removed double postings
I guess people have suggested a feature to change lyrics on a song, instead of extending it?
For people who like music but don't feel creative (many!) it would be nice to have a "personal artist" feature. That is, a service that explores some data from an user (for instance, their YouTube history) and maybe other social media and comes up every day with a few songs tailored to the themes and genres of that user.
The easiest (and most profitable) way would be integrating it with an on-line music platform (such as Spotify or Google Music), so no outer scanning would be needed.
Instead of creating an unnamed artist, the service would also create an IA-gen artist (band or solo artist). Apart from making the music more personal, it also would be easier for the user to specify "please go on with these songs" or "never want to hear it again".
As tailored-songs, it would be nice for the users to share the songs with other people, knowing that it is something special that nobody has hear before.
According to my experience, even people who like music are afraid of use Suno, even as easy as it is for 2 main reasons:
- They like music, but they don't know exactly WHAT they like.
- They fear of revealing too much about their personal interests.
The personal artist solves both issues and it could be a good income generator, obtained by agreements with platforms that already have huge incomes.
Sorry for so many words but it is something that I do (waking every morning to hear custom music) and I think that could expand the user base of Suno!
Anyway, thanks Suno for making a dream of mine come true!!!
Perhaps working on a sort of DAW that lets you prompt each stem of the audio and then composite it afterwards would be interesting.
Like Audacity but run by Suno.AI model tracks!
that way one could influence a style better and cheaper!
I got a response that they are working on a Stem feature.
i'd love for suno to be able to sample from any point of an uploaded sound of my choosing! for example, i upload a 60 second sample that I want used, but I don't want 60 seconds of talking, guitar playing, or whatever at the beginning of my song. i'm not sure what the right words are to use to describe what i'm going for.
use audacity edit it down then upload its super easy
i know that already, i'd like suno to be able to sample from the audio that we upload, rather than just extending from the audio, is what i'm saying
oh i see what your saying now my bad
yea right now, if you get whole song from extension... it'd also include the sample at the beginning of the song... wish they did not include it at the start of the song
its really early im guessing its just a proof of concept to play around with showing the potential of what it can do
yeah still a cool feature for sure
My suggestion is a tab on the website for a leaderboards for the summer of suno contest!
Not sure if this has been suggested or if something simiar has, but I'd love to be able to customize a song by giving it a specific chord progression like C to eMaj7 to D to Cmin etc...
to add the audio prompt to free user but with a limit of 3 to 5 uses
increase to 100 credits in the free plan
@dire frost Checking music2music That's amazing. Uploaded 2 songs and check the difference:
- First: 59 secs training of my old 2019 recording (4 min, I've cut the first part: 59 secs) on 2 iPhones (backing guitar, solo guitar). That's no quality and rythm, but always recording just to remember the melody I made that time. Forward to 59 sec to hear generated next part (made female&male duo, loving it). Unfortunately keeps strictly style. No possibility to add style by style prompt nor [] tags, keeps the style. Or I didnt manage last 10 tries.
https://suno.com/song/000b079d-655e-4a1e-94bb-36e5f7c58ee7 - Second one: Same audio input, but extended from 0:03. Generation starts from 0:03. Easy-breezy to change style. But the melody has almost no connection to 3 seconds of music I trained, almost no connection to melody in all audio file. My thought: that first audio2audio version is a tiny toy, to upload first short sample and include it in music, like text prompt. The real gamechanger would be toupload a recording and make Suno to sing it, change style, extend inside if it would fix, learn melody instead of instruments/genres.
https://suno.com/song/a5da3020-076d-408b-b42f-a3dab490db9d
add: piano, drums, song. Listen and make your own with Suno.
rock, drum, futuristic, romantic song. Listen and make your own with Suno.
Again, thanks for sharing
It would be helpful if we could direct Suno a designated time length on a final clip to end a song or to force an ending via a new metatag.
It would be really cool if you could save a voice from a song and be able to select any of your saved voices for other songs.
Would be great if we can get a tier of payment for unlimited generations
Also having a verified metatag list along with a suggestion list for new metags would be great, Not sure if that can be done or if the metatags are immutable at this point. I woudl think they should be able to be updated but not sure what is going on under the hood
I'm getting here now and I believe my suggestion may have already been mentioned, but being able to upload the audio of the a cappela voice, have auto tuning, and add instruments would be incredible. It's almost what you can do with the new system, but that would be more specific
never mind, I just did it now lol.. Gsus.. What you have done
It would be nice if it was clearer that the songs are busy generating using indicators for example, now after pressing generating I didn't immediatly noticed that my songs weren't fully generated empty time length isn't enough imo
I have a very nice song created by suno but the bass is vibrating too much, too many intererences in the bass. Can you please include a repair function for songs?
What's more, there are times when SUNO generates a very good and contagious style of musical RHYTHM, when I give him the instruction. But this is no longer played in another version to create or when improving the lyrics. For example, I have a music A that I liked, but I didn't like something in the lyrics and I changed it to generate music B this generated no longer comes out the same or something very similar to music A. It would be great if there was an option where the musical RHYTHM generated can be chimed with the change we want.
(of course to avoid a lot of copying and pasting from other users, this option would be enabled so that the user who creates, can apply this function but not when someone copies the promt that is if it is an option already in operation in SUNO) @echo canopy @open rock What will he think @strong swan ?
Don't know how the economics of these models work, but it would kind of nice if you could pay fewer credits for shorter songs (like 30 seconds), which you can use to experiment with sounds before extending them using more credits into full songs.
I had a really cool song generating but then it suddenly disappeared on me. Is there any way it can be retrieved?
It was not in my deleted folder.
In your library, do you have either the liked or public buttons turned on?
Nope.
Piggybacking off of the bug above, a simple button for "copy style" for a song would be nice
As on mobile it is tortuous to copy
due to them now being links
the workaround is simply "reuse", and then repasting your lyrics over
If it isn't just on a different page, I don't know what would make them disappear
But this would be better
While it was generating I was clicking through the other menu options and only the second version (not as good in my opinion) ended up being the only one remaining. The credits are now at 1705 instead of a normal credit deduction.
That sounds like a generation failed and was refunded
Is it server side or did I do something to mess it up?
I would guess it was server side.
I mean, it just sucks because I could definitely hear a portion of the song and really liked it.
Not sure where to put this but a big thumbs up to the Suno team for officially becoming a true tool for music producers with the Upload Audio addition. I tried Udio's feature and it just added the new directly to the end of the upload which wasn't great. Seems that your solution redoes the upload and includes that in the full mix. Brilliant and how it should be. Keep up the great work guys.
I miss the remix option
It was renamed to "Reuse Prompt" but does the same thing.
Idk if the suggestion was made, but it would be great that we can upload a song and the IA analyze it to recup the same genre, structure and rythm of the song and generate one similar
It does not, it used to have a similar vibe but different. Now it’s not even a similar vibe just a whole new song
Yea thats the problem i am facing at, cause i've made a song with a "EDM Variation" prompt and can't find the same vibe
I am so desesperate
@open rock
Or no credits limit
it did not... it always just copied that's why it got renamed in the first place
@nova quarry That is a good suggestion. If there was a slider for the upload creativity. Full is the same, less is something similar.
Yes like a stable diffusion
Or even using negative prompt
that could be great
Why when you share a song, or download the video of it, it shows all the AI generated pictures to it? Why not just give us the ability to just show the Cover art?
It seems that The AI has problems when creating songs in Spanish, that would be a point to correct.
The new update is very cool though, well worth the money
I am having problems making classical jazz crossovers. It makes it only classical or jazz but it can't do both.
Yooooo if y’all want any song recommendations y’all go listen to put it in sport by cool coesta cause bro is tooo 🔥🔥🔥
Move the upload audio to inside create, I’ll never use it and don’t like to scroll horizontally to change pages using my phone
That isn't what this channel is for
Please improve the sound quality of the mp3s so that they can be downloaded as wavs instead of converting them to wavs!
there's already an option to download as wav
Please improve the sound quality now! Instead of a specification where you can convert music composed in mp3 to wav and download it, please make it a structure that composes in wav from the beginning! This is proven by the fact that even if you download a wav, the frequencies after 16khz are cut, just like mp3!
No one wants converted wav
oh you want separate stems? it's already one of the most requested features. Some day.
I would like higher quality audio as well. However, the WAV downloads are not converted from MP3. Here is a frequency analysis of one I downloaded yesterday which has frequencies up to 20KHz present. Audio quality can vary depending on what gets generated. I have had some songs generate that lack high frequencies.
The overall audio quality is limited by the model. Audio quality should improve as newer models are developed.
Well, that's one of the requests, but what I'm pointing out now is whether Suno AI composes music in WAV format from the beginning, such as 24bit/48khz, at a music industry standard rate. This is different from the specifications that allow you to download low-rate or mp3 music by converting it to wav.
So how do you explain these? The frequencies played back are the same part of the same song, but the one on the left is mp3, and the one on the right is wav, but in both cases, frequencies above 16kHz are cut off. Indeed, mp3 is completely missing, and wav is compressed to less than 3/1 of normal compression. In any case, it is impossible for WAV to visualize such frequencies.
I don't know the technical details of the models so I can't explain why the high frequencies are lower. The developers said with the current models that WAV would not offer much quality benefit due to the quality being limited by themodels. There is a little more detail in the WAV, so they are at least not converting mp3 > wav internally.
The only way the quality will improve is with newer models (v4 is being worked on.)
Even if this were the case, the sound quality would still be for amateurs, since the results were obtained even with V3.5, which is the latest model at this point in time. Most of the cutoff in the high frequencies, especially after 16-18khz, is due to compression, and even if the mp3s are not converted to wav, the lack of detail is a problem.
auto cutting of the clips would be great
Please make us able to to unlink generations, so that we can get rid of the original vocal sample we used? Also, it would be great if we could have a system where we could apply to send in samles of our voices and have them "cleared" so that we could use our own voices on these Suno songs? ^^ Thank you for reading
yeah hope theres a function to remove vocals on completed projects
Please please add this
A small feature, but what about the ability to put gifs in the Cover Image section?
can the create page stop calling focus to the bottom of the list and the newest generation?
that's a feature that serves no purpose, and just wastes the users time, like the clickable links in prompts
I want a "cut song" function to delete the excessive part from a soundtrack, because the v3.5 sometimes over-generate excessive length if it thinks your track is too short.
Like if all my lyrics ended in 2:52 I can cut off anything later than it and no more repeating.
Over-generation is ruining good tracks 😭
suggestion: lets do crowdfunding to buy some good developers or some good managers for at least the user interface ! It is unbelievable how many weeks can flow without a SINGLE change! And this for things one could do in few hours or few day.
I think they cut over 16 khz on purpose, so the songs stay in "radio quality" ...it is not the "AI model", it is a filter and it is strategy!
Not sure if it's already been suggested but it would be pretty awesome to also be able to edit the lyrics section of our songs page. That way the presentation can be cleaner and be made to reflect the actual generated lyrics. Assuming there were errors or something got omitted and to clean up prompt clutter.
Without a single change! Like how we didn't get the search function. Or song radio. Or the ability to upload reference audio. Not a single one!
It's not like we got a new feature literally today or anything!
I bought a $30 monthly version, hoping that my music can be sold or commercialized. Where can I download the copyright certificate to prove that my music can be commercialized? After I uploaded the music to the music website, the music website failed to pass the review. I need to upload the authorization file of the ai creation platform or the voice authorization of the ai performer. What should I do? Is there a special authorization document to solve this problem? Please give me an official reply from suno official. Thank you.
in browser song cutting option, fast and easy of adding notes to your tracks, male/female voice options that works, thanks
I don't think there is such a thing, from everything I have read its just a given that as a subscriber you can commercially sell your creations, I will certainly keep an eye on your post as it would be good to officially know for sure.
#💡┃suggestions-web message
Repost 👆
The new feature make me think that Suno goes into the wrong direction.
We need to provide the music that we want that is a good point.
But i would like to use the functionnality by providing only the melody and chord progression. and would like suno to analyse that and create a full track based on that input.
Not to provide an input and make him to invent the next part based on the first part... that's not a good approach.
So we should be able to import just a melody input + a chord progression input. and after analysis, he can create a new song with different style by keeping the main melody and chord progression as heart.
Sorry If already covered : 1. I often get 'overplay / extra generated audio on songs. A wave form style ' trim' tool just to tidy up the start and end of songs that are up on my profile would be great. (I already trimmed the downloaded version so I guess I could now upload those and use / extended them but that would end up degrading quality etc, plus I'd have to use credits for something that should ideally already be a feature.
-
A search bar for the 'explore' page. And the ability to manually rotate the ring. Waiting ages for it to come around to find / re-find a genre to listen to is frustrating.
-
Ability to keep / reuse the seed of a particular vocalist. Every time I regenerate I lose the voice I really liked previously.
-
Ability to 'preview' even 3-5 seconds of what the song will generate as. I go through about 20 generated songs before finding the pace / style I like.
(Although I guess that's how you make money. I must have spent $200 just regenerating the same 10 songs until it was 'just right'.
Thanks guys.
P.s. The trim tool would be great to let me 'extend' from a particular point rather than just the garbled ending.
4.2 : a small scrollable list window to 'sample' explore genres next to the generation window. When you change page focus to go to the 'explore' page, all the input in the generator gets wiped. ( Plus I forget half the exact names of genres). Having a 'sample style' box next to the input section woukd be great.
- Vocal emphasis modifiers ( sort of like Amazons TTS script) would be nice, though oi guess partly defeating the point of an AI voice. I notice some, like caps, parentheses , exclamation mark, but often the lyric stress points are wrong or timed/ split badly despite punctuation.
Cheers 🙂
I think for me right now instead of adding in more features I'd rather have what we already have fixed, like the really terrible audio quality since v3.5. no point being able to add our own audio if the overall created output sounds like trash. its just distortion, distortion, distortion and really low wishy washy audio. A lot of people have been posting the same comment and some posting audio spectrum graphs. I have wasted now nearly 2000 credits in a day just deleting song after song of what could have been fantastic creations if they weren't ruined by the distortion and wishy washy sound. Please address this Suno Devs. I'll buy you all a large Big Mac meal with a McFlurry for afters 🙂 🙂 🙂
If this has not been said yet, it would be nice to be able to listen to what you have done so far of the song you are actually making. I mean I'm actually at part 6 of a song and I would like to hear the result yet without "Get whole song". If I listen to all different parts, I don't know exactly how long it is because I often extend from the middle of each part. 🙏
Suggestion: Add vocals to song. So i upload a piano composition and then i can add other stuff like vocals
Be able to edit a song that we made on there. For example, if we have a perfect ending in the song but for some reason it kept adding after that point we can edit it to delete the extra.
yes inpainting outpainting
Would be absolutely amazing!
I know this seems like a minor thing, but the song upload length to 60 seconds rejects tracks that are 60.0 seconds, which is everything that v3 generates, making for an unnecessary amount of extra editing
improve the sound quality of the mp3s its bad rightnow
has anyone requested the ability for subscribed members to purchase the rights they made to songs while they were using a free account? i would happily pay extra money for the rights to my favorite songs once i get a subscription.
I'd love to have a re-sing feature, to be able to edit the lyrics of a song I've already made on Suno. I've made very good song but I always end up improving the setting over each listen session I do but I can't edit because the extend feature might just ruin the end of the song (changing the structure for ex)
On top of the edit I mentioned, maybe also have a download audio only so you could use it in open verse settings too.
I'm not too fussed about the audio quality. The day I just use AI software to produce a finished work is the day I... Mmm..I don't know...something horrible...start drinking Pepsi instead of Coke... I write all my own lyrics and re-re-re edit them until I get them to suit. I would consider the music and vocals a great 'strong guide and influence' to create the actual piece manually. I do the same with stable diffusion ai art. It's for ideas, inspiration, knocking out demos. It's an unpopular opinion but I think the mass of button mashers than have swarmed onto AI, so proud of what "THEY" created, is puke worthy.
Hi, I am so excited about the audio upload feature! It opens up a world of possibilities. One small feedback: it would be cool if you could upload longer audio files and crop them. I can cut them in my DAW but it would make the workflow easier. Thank you so much for the new feature❤️
It'd also be nice to be able to have the option to exclude the original sample when expanding and merging clips together with "Get Whole Song".
I have a sample that I used to see if I could get some control over the range of the singer (definitely worked) but the song ran over and if I try to force an ending with v3 expand/[End] trick, it slaps that sample at the beginning. That sample will (hopefully) never see the light of day. 
It's easy enough to trim that off in post - just makes it un-sharable on Suno.
Ok, this is obviously not a high priority, but this morning, for the first time I was trying to listen to a user playlist and thought it was pretty decent. But when I'm listening to music at home through Spotify, etc. I'm listening on my Sonos speaker. Any chance of a Sonos app for listening to Suno? Obviously only the bored person on the team can look at this, but if long-term Suno want people sitting around getting their music from the website, there's worse ideas. 🙂
start drinking Pepsi instead of Coke
WTF!!? NEVAR!! 
Blasphemer!
PortaBob: I noticed a similar thing. They song sounds great on airpod 2 , but total garbage on almost every other speaker and headphones I tried them on. And I mean THX certified, Bose etc. Only with some filters and EQ to reduce bass etc did they sound ok
Suno Tags, Please!
The, Upload Audio is a major leap forward. Amazing results. Suno is now a song writting companion. What I would like now, is to be able to interact with the song. To do this we need, Suno Tags. Structural Tags that we can place in the text and directly impact on the song and lyrics.
We all play around with Lyric Tags and get varying results, dependent on the Style of Music being used. But I'm thinking, Suno Tags would be universal and work with all Styles of Music.
Suno Tags, Please!
Keep up the great work you're doing!
Is it just me or did basically every suggestion I just made get repeated pretty much directly afterwards? Ah well
yeah.. 😉
Mine might have been bad, but it appears to have been original. 😛
A bit niche though PortaBob
Absolutely. But thems what don't ask definitely don't get.
Yeah they do. It's called 'taking / stealing' :p
Just a couple of suggestions , I think a useful implementation for your service could be to allow you to intervene on an already generated song from our private library and modify the lyrics so that the A.I. can keep the mood and the structure of the song and make some changes without creating totally another song.
Another functionality really useful could simplify the previous request could be that when you choose only from your private library a song and say "Reuse Prompt" the A.I. evaluate also the track and not only the parameters as "style" and "Lyrics" but also the source song from you choose to reuse the prompt.
Inpaint it?
Finally it could be useful to extend a track not only from the end but also from the beginning for example if you want ad an intro e set the join the specific time
This is seems to be closer to AI art stable diffusion generation than most people realise. Most seem to see it as totally separate thing, but it's not. So if you're familiar with aisd try thinking in those terms when suggesting / having ideas
Sorry Alice did not catch, I only suggested some feature that could help to develop a song with more control.
samael: well. It was unclear what you define as 'extend'. Because you could just generate an end extend and paste it on the front. And Whalla, extendiend. If you mean created to merge seamlessly / possibly modify the original to suit, then you need to regenerate the original + intro which changes the seed and you end up with..not the same thing...
Never mind. Not enough coffee yet
Depends how they are doing it. I just meant (a)+(b) is different from (a +b)
Oh yes... you've been more technical ^_^ ,
in fact, I tend to think more about the scope and function under the hood 😅
I haven't read the coding papers yet so I'm bs speculating... :p
I'm not an AI expert 😅
I'm just the lazy user playing with a black box without knowing how it works🤣
Is there a cat in the box? 😉
and is it alive or not ? 😉
Schrödingers cat? 😹
and if a tree falls in a forest and no one hears it fall, does the tree exist? 🤔
Perspectively: does the tree fall, or the forest?
would be so nice if u could do now with the audio upload that the ai can do only normal talking too and not only music
One thing I'd like is a feature that would allow us to describe the style of a generated song from Suno.
My case is, I often try to generate random thing with small prompts like "humorous, metal, deep voices". But the generated song goes beyond that and add it's pinch of salt to that to make the final result. And sometime the result is pretty good, but generating again gives something entierely different due to the prompt being pretty vague. I'd love to take the generated song I liked and be able to know how I could recreate something similar by enriching the original prompt.
TLDR: a describe song as a Suno style of music prompt feature.
It could be useful to get the seed from a previous song. And so, use a specific seed as optional parameter to generate a new song.
What about making "Remixes" - for example - to change the reggaeton song to HOUSE but keeping the structure and the whole concept of the track?
experimental theatre
try speaking for longer, cleaning the WAV up (noise reduction etc), and this prompt
I thought there was a 2 min limit on songs so wrote to that but I get randomly 3 mins, 4 mins. 3mins 38 seconds..just randomly. Makes it hard to write to
slightly off topic, but I feel important to mention : Honey is delicious on/ with natural Greek Yoghurt. Maple Syrup, not so much. 🤮
( A public service announcement brought to you by DarkFae Enterprises)
Slightly 😆
There's an #🗣️┃chill-chat channel, which is strangely on-topic a lot but I'd hope the devs don't see this here and start mixing honey and greek yoghurt in with the audio. (If they do and it makes worse audio quality, we know who to blame!)
Oh you bet 🙂 when I go REALLY off topic, the fabric of the universe starts to unravel...
✅ 💜
Love the upload audio, And not sure if its needed but there should be a 'lyrics text' or a "what is the audio" to help the generator to understand what the audio is. It could help if its "lyrics" "hooks" or just random "fun sound" 😄
PLEASE ADD A STEMS OPTION so we can get stems from the songs we create (Plz and Thank you)
stems don't sound that great now to be honest
is that okay
the sound isn't generated as stems, is it? so getting stems will probably always mean some sort of separation process?
If they change the model a little bit and make the stems be generated separately or separate them with audio source separation models,first would take a lot of resources and time and second is not really a good option since the current stems are not clear at all
And let's say the audio quality have gone to Hi-res someday,i know most of you will use vocalremover.org to separate those 😅
@formal bear I would settle for clear audio, more so since v3.5 so many songs are destroyed by distortion, distortion, distortion it sounds really bad, nearly 2000 credits in 24 hours all been binned because of this, its like so much of the song elements is severely overdriven and just creates a crunchy fuzzy low quality mess of the song. 🙂
also, everything turns to pop so easily with v3.5... as if they dataset is overtrained on pop
that, and the fuzzy vocals, seem like the two major things that need improvement IMHO.
now that we have .wav export, stem separation can be done using 3rd party software
Yeah like xminus or mvsep,if local then UVR5
exactly. they exist, so no reason for suno to re-invent the same thing.
Yeah, just gotta improve the quality
also some sort of folder structure would be nice so you could move songs out of the create screen in to user created folders like "Bangers", "Good Lyrics", "Ideas", "Instrumentals", "Songs" etc, the create screen just gets so cluttered. Sure there are playlists but this doesn't take the song out of create screen and you end up with pages and pages of stuff.
ok sounds like we should hold stems for now
Please don't bother devs with things like that. For questions like that you can contact the supportdesk. support@suno.com
Tony is the Dev that is in charge of the model..
sorry I thought this was a suggestions page? I dont know who Tony is or what he does so again apologies...
You pinged him.. suggestions are fine! If you want help with your account, you can email support 🙂
I did email support and 3 days later I got a screenshot and reply saying they don't get carried and I lost out on $30.00 which wasn't cool yet looking at my screenshot thats how it looks. Some things need to be clearer thats all.. Its not nice supporting something but losing out for supporting it because of something that wasnt obvious
And the suggestion wasn't the issue being pointed out, it was pinging someone specific with it
I guess we differ in that. For me it was obvious without even reading the text below. But yes, maybe they should be even more obvious.
The page looks the same regardless of monitor/browser size btw. I scrolled down to see the rest of the information about what the plans included and thought about it for a while before deciding which plan to go for.
If they moved the text from the bottom to the top of the page, I'd be fairly confident that there would still be people who don't read it. Even if it was a popup notification as you subscribe, people would hit "OK" or "agree" like they already do for other things (see: "custom lyrics" agreement 😄 )
appreciate that 🙂
yes I can totally understand that too 🙂
if you move a song to trash thats still generating it should cancel the generation. i do this alot where i forgot to change something so i trash it
This I'm definitely on board with. The more I use Suno, the more it feels like I need some kind of foldering system for organization as my "work space" is cluttered with multiple ideas on the go.
totally 🙂
It's possible but still not perfect... still muddy on the spot where vocals used to be... I can't wait for the days Suno generated them stems separately... as I really want to create some good Karaoke tracks of the songs
Instead of extending from an audio clip of your voice it would be nice if there was some thing that made you read a few sentences so it can save it as a profile and you can select which "voice" or profile you want to use when making a custom song
True, but inferential stem splitting is no substitute for generating separate stems from the start.
Suggestion
It would be really cool to be able to take a very short sound byte (maybe 2-5 seconds of something like a squeaky door or something), upload that, and assign that to a specific meta tag in a song. This way when you use the meta tag it will incorporate the recorded sound byte.
Example: I record the sound of a squeaky door. I upload the sound as a 3 second sound byte. I give this byte the meta tag name [squeaky door]. This byte is stored as a "song" for me but when I reference it in other songs with that meta tag then SUNO grabs this sound and lets me use this.
This would be great for when people want things like whispers, bells, whale sounds, etc.
It would be fantastic if a song was tagged as "custom lyrics", "AI generated lyrics", or "hybrid".
Suggestion - allow uploading of longer audio (5 minutes would be awesome, but i'm a dreamer)
Once uploaded, allow the user to highlight sections of the audio to be replaced by Suno.
So I could make an instrumental track, upload the whole thing, highlight the verses, and have suno replace them.
The ability to archive versions of a song that didn't turn out the way I liked (say a typo in the lyrics or something) and don't want cluttering my song list, but I don't necessarily want to trash because it's still useful as an example of a prompt
Changing the image for a song part should change it in the video that's generated when you get a full song
Would be cool if there could be more dynamics in the space, there's some parts that get generated that are panned or in the background. But it's so far center and forward for a majority of everything, could be cool to have a way to have a part that's specifically panned a certain amount
Since we are all on LIMITED plans, it would be amazing if we had the ability to select between ONE and TWO outputs, instead of being forced to waste the credits on two outputs with every iteration. The subscription says 2500 credits or 500 songs, but in reality it's 250 songs and their closest variants, which is oftentimes useless. I should decide if I want to change the prompt after I listen to the song! Why should I always have to pay with my credits for the extra song, that I don't want to have? Give thumbs up to this, maybe it will be taken into consideration, although I seriously doubt it :/
I agree with being able to change the amount of generations, but for the opposite reason
I want more
speaking of, since subscription credits do not roll over, and the fact that generating two songs at a time is required, if you have 5 credits left over you cannot generate anything, and it will not roll over.
that usually happens if one of the generations if refunded due to silence, etc.
yes, maybe in your case you want more outputs for the same prompt... in my case, I'm more than satisfied with one output per prompt. Both perspectives lead to the necessity of being able to edit the number of outputs OURSELVES 😄
Has "Thumbs down stops song generation" been considered yet? 100% of the time I thumbs-down a song, I'm OK with not hearing it again. It might as well also go to the trash, too.
Your devops team and your finance team would probably love if this was the default case.
I don't think the finance team would love this 😄 Instead of using 100% computational power and time for each input, they are using, probably, 125-150% to generate 2 songs with the same input. Compare this to using 200% for 2 songs. They have a profit of 50-75% computational power and time for every input, and 100% of credit costs for every input....
I strongly doubt the financial team would like it, since it doesn't charge the credits for generation until the song successfully generates to completion. Being able to cancel the generation would make them do the work of generating part of the song but never paying them for it.
On a free account, I could probably review the first ten to thirty seconds of every generation and then cancel before they complete, effectively giving infinite credits unless I find a near perfect song that I want to keep.
Credits are independent of this equation. They still generate 2 songs and charge the same credits. It's about resource allocation.
Current: The computational cost per song stays the same.
Proposed: Obvious computational cost savings if they stopped jobs early.
<The ratio of thumbs-down to kept> *
<ratio of the song's completeness > *
<the current computational cost per song>
= reduced computational cost per song.
And if they were throwing them in the trash by default, it's an obvious cost savings on storage.
I expect storage is extremely cheap compared to compute time.
I was originally annoyed i have a clogged up library of hundreds of songs that belong in the trash, and that some really bad songs keep on generating.
What is going on with 3.5?nearly every track is around 1 min to 1:20 max ,its meant to be up to 4 minutes. Any time i do get 4 mins tracks they stop half way through and start up again after about 5 seconds, come on seriously! why cant we have an option for 4 minutes manual selection and also an option to manual select an to entend to an ending because in build 3 i sometimes get tracks going up to 10 minutes with no end in sight. Also why is description limited to 120 letters, but lyrics can have hundreds.
Can you link one? Most people have issues with it going to 4 minutes no matter what, so I'm really curious.
Suggestion
You can copy a song's overall style, instrumental style, vocal tone, etc to use as a base for another song, similar to the similarity of a song extension once you extend it, but instead this can be used for a new song
This has probably been suggested before and shot down but I wan't to say it anyway. Mostly for the people who intermittently subscribe to the premier plan due to money constraints and can only choose 1 or 2 services. Maybe let them keep what they've paid for and not take away just over 5k credits or however much the have left when the subscription ends.
Are you asking me to link one? All my music is private so im not able to link,90% of my tracks in the last few days have all been below2 minutes on 3.5 and those that make it to 4 min which is about 0.5% they all fade out mid track and then restart again.
Yeah, I was hoping to see what was happening, since it's so different from what everyone else complains about. Most people have trouble getting a song to stop before the limit, so someone saying it's stopping really short is interesting.
You can basically already do that, you just have to be good at prompting and know exactly what you want
Ability to perm delete all in the trash bin than single selecting!
Found a quicker way thankfully
can we buy Suno stock option soon ?
Guys. I’m sure this has been discussed one or a billion times before.
But:
Stem separation and .wav / whatever lossless file format. Or at the very least 320 mp3.
If you implement those two things BEFORE some of the half-assed competitors you’ve currently got, then the vast majority of the music industry will come running.
You guys could literally add a $299 / month GodMode-tier for accessing those functions and no one would question that.
I’ve been in the audio engineering / producing / writing business for just over 20 years and I’ve never felt so good about from a business / market point of view it as I do now. AI won’t take (all of) my jobs. You’re literally about to open up a door to INFINITE possibility and creativity.
Not sure if this wall of text qualifies as a suggestion. Sorry for the spam lol.
Kinda just wanted to remind you guys that you and the tool you’re building still have the most epic glory days ahead of you!
Thank you and please keep it up! 🤝💯
Wav was added for paying users on 5/24
one of my songs was deleted, I don`t know why? Please undo it!
I payd for it!!
When you are looking for it in your library, do you have the liked or public buttons turned on? That gets people sometimes.
120 characters in the "style of music" field is very little.
nobody deserve!
How about doubling that amount, suno.com?
Where is my Song "Genug"?
I payed for it!
Where is my song???
no, there is nothing
Its just gone, and I dont know why
Please tell me what happened. please tell me.
I payed for it. I want my song back!
It was nearly 100 bucks!
I can't find my song in my library! What do I do?
- If you currently have the "infinite" loading when trying to access the library, please see #💬┃general-chat message
First, reload the page and ensure you are in the library. For cache-bypass reload, see #📚┃suno-school message
For usual causes:
- Could be on the next page (blue circle for pagination)
- A filter is enabled (yellow circle)
- Accidentally trashed (red arrow to access the trash bin)
- You are in the create screen, not library, and the piece was down voted (which causes the entry to be hidden)
- Accidentally overlooked the entry (use the text filter option). May need to reload the page first.
- Logged out or logged in to another account (can double check by going to your profile)
https://media.discordapp.net/attachments/1102676217687195679/1246851079182880808/image0.jpg
Dont fool me - I know. It is not in my trash. It is gone.
bump was literally about to suggest this, stems would be extremely useful for producers like me
If you do not see it when using the text filter option, do not see it in the trash, made sure you are logged into the right account, and basically none of the items on the list helped solve it...
Not sure what else you can do other than email support@suno.com with all associated information (e.g. your account email, profile name, the name of the song, when you approx. generated, what version you used, etc.)
It will take a few business days for support to reply back to you.
Ok, i checked it out. My Song is gone. Thats true, it is not there anymore, it ist gone and lost and NOT THERE!
Contact support@suno.com with all relevant information then. You will likely get a response sometime next week (since tomorrow is Friday and likely the waiting list for responses and investigation time will take more than one business day)
Thank you. I`ll ask on Monday.
Is my Song that important? Its just about everyone being equal. is that a crime?
AND its in German - I mean - who can speak german ?
Fashist Germans are not bocked or tagged - Why am I? I'm mot fashist or racist or anything!
*blocked
What did I do?
It's more than likely been said before but, trimming beginning and end of songs could make a major difference especially when using the new audio upload feature. Also a mass deletion option for your trash.
It occurred to me that a primitive algorithm for lyrical quality~similarity weighting could look for lowercase solo i's for radio stations for songs which contain them, since GPT, unless explicitly prompted to do so, doesn't use lowercase i's (lots of human don't too, but I've found more than one might think)
there are likely other idiosyncraticies of human written lyrics that could be similarly used, if that's something desirable (the entire concept of lyrical weight may itself be flawed, just an idea)
I bought a $30 monthly version, hoping that my music can be sold or commercialized. Where can I download the copyright certificate to prove that my music can be commercialized? After I uploaded the music to the music website, the music website failed to pass the review. I need to upload the authorization file of the ai creation platform or the voice authorization of the ai performer. What should I do? Is there a special authorization document to solve this problem? Please give me an official reply from suno official. Thank you.
I am amazed at the speed with great new features! Here's a suggestion: Train the AI to create jingles and DJ drops. That feature would alone encourage me to upgrade to a paid plan. DJs and small internet radio stations are always needing fresh drops, jingles and promos.
which distro did you use to upload your songs
The new "Audio Inputs" feature is the bomb. But it would be much more convenient to add a filter to the Library, by which you can find sounds that you have uploaded to suno 💞
Make it possible to create a whole song, so that when I cut out part 1, the other parts would shift and part 2 would become part 1 instead of the deleted part. It's quite useful now when I want to add just a specific voice that I want in the song and, after finishing the song, delete that voice so the song doesn't start with it
As an owner, composer and producer in my studio I use Sumo Ai. However, I still don't understand why the quality of the singers is substandard. Despite having two paid accounts, they are virtually useless. How come ElevenLabs manages to achieve what Sumo Ai cannot? It almost feels like there is manipulation going on. The dynamic range of the voices is not feasible; it's quite disappointing because Sumo Ai does quite well in terms of compositions. I fervently hope for improvements; otherwise I will consider not renewing my accounts. Conversely, Udio performs significantly better with their singers and vocal quality.
“With 45 years of experience in the music industry, I have witnessed countless changes, but this is a groundbreaking new high. Join us as we embrace the future of music and set new standards.
Hi Suno, as a professional artist I love using Suno. But I have a couple suggestions for you to make your app perfect. The problem now is when you already made a song you cannot change the lyrics after and that is really bothering. Sometimes you have the perfect song but a couple of words are just wrong. it would be nice if one can change the lyrics after you made the song. Also please make it possible to make a copy of your song so that you can make another version for example in a different language but using the same song. Thank you so much.
Can we have an Extend (without connecting?) ?
Reasoning... we stripped out the vocalist from another suno song of my own: https://suno.com/song/91340c06-643b-4446-bcd3-46122e7e4840
A song. Listen and make your own with Suno.
Then made this (example): https://suno.com/song/f0b6ec1a-1b63-467a-a859-b7d48602d56a
Epic, Heavy Power Metal, Symphonic Battle, Electric Guitars, Bass Guitar, Synthesizers, Ambient Ethereal Cat song. Listen and make your own with Suno.
I'd like the start of the song to be from after the sample realistically
which needs a 'delete all before' or something....
Creating a realistic flow to reuse our favourite Suno AI generated vocalist for more tracks....
Hah. I came to post the same thing. I've made some recordings of myself making nonsense vocalizations in an attempt to get some unique character in singers' voices. Seems to be working quite well, but every single song runs over and I can't extend them without pulling that original sample in. That ends up turning every song into a comedic horror 
Is it possible to pay with paypal or goole pay?
Make a song with a simular style, often there is a lot of difference between the songs even if you set same prompt (and I understand why). so please add a feature to make a simular song (for example as a button near the extend button)
I'd like to make "whole songs" with just part2 and part3 to leave out the input audio
Also, I'd really like to share my audio upload songs and make them public, they are some of the best I've done...
Suggestion: I would love it if when you make a Custom song. And leave "Style of Music" blank for random it still says what kind of style it picked
would be great for my youtube vids 😄
alright I figured out the bpm part now but can we have bpm fades.
since overtime bpm seems to be a luck-based prompt at this point
i wanted to suggest a simple feature for the create tab. "clear history" to reduce clutter and start a fresh session. maybe this helps suno's memory? (this doesn't actually delete the tracks) I am not sure. I am looking at this more on the side of "less clutter".

hey there... you guys really would benefit from letting the user define the tempo of the output if they want to... over-wise the tempo is not exact and therefore can not be used without time-streching which messes with the sound even more..
sometimes putting [tempo insert tempo] works just fine. depends on uploaded files I guess, but it usually dictates it.
Also you should have a clickable box that allows the user to discard the uploaded audio from the output-ed song... (like in Kaiber it has a box that says (Display picture in first frame of video)
' [tempo insert tempo] ' wow.. that would have been handy to know that yesterday... 😉
this I am not sure
could be cool
not sure how it would be coded, but nothing is impossible
a rough editing feature so that you could cut together your song into sections before 'getting whole song'
agree on this, it would make it a lot easier to understand how to cater to it, or at least understand when you just don't have the numbers you think you need. Like, this, based on prior experience should make it on there, and I made sure to remove anything inappropriate
accoustic anti-folk, raw-pitch-steel string, hoarse rough delta hyperblues, amateur underproduced session lalalalalaaaaa song. Listen and make your own with Suno.
and didn't change the cover
so figuring out if it's just a busy time and isn't hitting the chart
or if it's just not allowed on
would be helpful. Also would help identify if some things are incorrectly not allowed
but this is like, middle America
salt of the earth
you could air it in Ohio and no kerchiefs would be raised in angst
they really should have a option regarding time limits now and also an option to extend but with the caveat of selecting extend but with the premise its to the end of song or track, Surely it cant be that hard to program. Version 3 always gives me full 2mins and the beta 3.5 regular gave me 4 mins. So frustrating
you want to take Suno to the next level? Figure out a way to separate tracks/instruments/vocals so we can change/regenerate individual aspects of a song. So many times I love a voice but not so much the melody or instruments and vice versa. keep it behind the payed plans. This would end any competition with other AI music generators.
How about generating music that doesn't sound like it's coming from a tin can telephone? That would help now. Or are everyone's ears broken these days?
you probably have, along with us, to wait years for a simple change a normal programmer can do in 1-2 hours...these sugesstions has been given already weeks before and still no move...
Being able to choose between getting only 1 generation instead of the standard two.
Alternatively being able to get a different set of lyrics and / or production of two generations is an absolute necessity 😋
keep dreaming....
how about not rushing to generate when I want to type ALT+ENTER or SHIFT+ENTER or CTRL+ENTER to go on the next line and instead considering that combo as a "SURE, I Don't care what you were doing there, we'll go ahead and create the song for you".... AS I am typing my lyrics
That’s what we’re here for 🙂↕️
It would be nice if the time info (extended from 00:**) was saved in every extended song, so the user can go back and just set the same time again. This because there are so many extended versions from so many different songs going on at the same time.
i'm probalby not the first one to point it out, but it's a bit disappointing that having PRO, I don't get 50 credit daily refills when I run out 😖 either need to pay more (and ordering addition credits is pricier) or wait until the end of the month. i'm almost thinking of getting an additional free account to use while i'm out of credits, doesn't feel right though.
50 or 100 credit everyday available for all but you gotta use them or they dont add up for the next day ... im a premier member too and having 0 credit left and 20 days to wait is LOL but I understand they are here to make money also cannot give everything for free ... but this limitation people will just create numbers of 'fake' user i mean 1 guys with 10 emails for 10 free account......
It sounds like they ARE planning on doing stems, but it also sounds like it's definitely on the backburner. 
exactly
It's never been on the backburner, but it goes "hand in hand" with improving quality to begin with. When they quality is good, then stems makes sense. They can ship it next week, but the quality of the end result will be poor. See this conversation from yesterday starting here: #💡┃suggestions-web message
Thanks. I was going by one of The Purples making a comment to the effect of "looks like we can put them on the backburner then..."
Gotcha.. well just pointing out it's absolutly not. They just want to do a good job, not rush it. The user experience is very important.
yup!
I think the Dev comment was due to people wanting stems being more technical / srs / doing post production AND already having tools to split out stems themselves - so the value added was questionable.
but I probably misinterpreted it.
make an android app to listen to the music with android auto support
Add a way to transfer your song to different model versions
Ex: Say I had a sing I mad3 with V3, and V3.5 recently got released. I can do something and put that exact song into the new model
Hello, first time writing in this channel.
I have two things to point out... but I don't know if had already been discussed, in case... my fault 🥲
- What about transforming this channel to a forum-like one? I have seen many discord servers using this system for feedback reasons. It will give moderators an easier life and also bring to top most discussed themes. Also... everyone can see what has been already pointed out or not
- There will be the possibility to have comments on suno.com songs? This will help to get feedback from other users without the need to post the track here on discord, if made public obv
Allow playlists to play on repeat, maybe it’s just me but as soon as it gets to the end of the playlist it stops playing anything.
I can repeat a single song just not a playlist
please stop giving suggestions in vain, NOTHING is being implemented! You'd better start some crowdfunding to find at least capable user interface programmers for SUNO.
come on, the 30 bucks for 2000 songs and for the amazing music the software delivers is not even peanuts, it is a quarter of a milligram of a peanut! I only complain about the user interface,
but in vain. The only UI programmer they had must be dead...
I can assure you that their programmers are very much alive.
not the ones for the UI !
can we please get a search filter in our library that lets us see our uploaded audio
I was wondering WHO gathers these suggestions here for SUNO dev, if there is one. Because NOTHING has been implemented in many weeks and most changes requires only a couple of hours...
Single time tokens puchracing, single month subscription or just an address for donation (some people enjoy the suno, ready to support it, but don't have that much money to get a subsctiption)
You can already make additional single purchases of credits once you have a subscription, or alternatively you could always do a custom subscription.
😩
contacting staff, planning a subscription etc just to donate $1-2 ?
dude
what UI do you want?
the app works fine
that's enough
build your own UI if you wanna something special my little dudy
It'd be nice to have a filter on my library for "Whole Songs". I want to be able to locate the complete ones that I forgot to add to a playlist.
Would also be nice if the downloaded MP3 prefilled the metadata like Title, Artist, and genre. bonus points for using the playlist I downloaded it from as the Album
these are my cd portraits for my songs of suno , pls make available for free users
It would be better and more practical to output, at least musical instruments, in a midi file rather than in a 24-bit .wav file, 48,000hz.
Yooooo if y’all want any song recommendations y’all go listen to put it in sport by cool coesta cause bro is tooo 🔥🔥🔥
would like the ability to share one song without sharing my full profile -- currently sharing a song gives people a link to all of my songs
are you guys creating stems as part of inference or are you doing stem separation afterwards?
Nothing implemented for many weeks? This feature was added 2 days ago: #📢┃announcements message
It would have to be something that takes place afterwards.
The Intro function does not work 9 out of 10 times. Most of the time the lyrics start right away, and thats pretty much unusable. I have tried many iterations of the [Intro] functions, but most of the times it does not respond. It costs me a lot of credits to get it right. It's like a lottery. Hope they can fix this.
What usually works for me is [Into] ... [Instrumental] ...
It seems to force it to slow down a tiny bit and realize what it's doing
Are there any subscribers who are willing to use a few credits to test an idea I've had regarding this? I think it might be possible to encourage it to use the same general sound for a song. If you use a song you've generated (or part of one at least) as an audio upload reference and give it the exact same prompt that the original was made from, does it generate something close to the same sound with variation or does it do a completely new thing with it?
Give me a refund when your generation decides to start to shout tags in the song. Lol I had the [falsetto] tag work in some songs and other times it shouts "Falsetto!" Same with stuff like [break] lol
Can I get a Fallout theme genre room? I have a playlist but the genre variety doesn't really suit any of the current rooms
Please create some sort of structure for reporting people blatatantly steal others work. Several on here were reporting this was done to them last night.
Im sure this has alrdy been suggested but it would be cool if we could keep the music and lyrics and beat and things but regenerate the voice used until we get a specific voice we like
Hi, is there someone I can contact regarding the terms of service?
what do you want to know? maybe we have a answer.
How about a flow option for famous songs that combines with other lyrics like lovely billie eilish, kind of like a remix.
Given that we aren't suppose to use music we don't have any copyrights to, that seems like it would violate the TOS and probably the law
Can we have the option to change the width of the browser text fields width wise, changing the length is great, but more options for that would rock and roll.
I have hardly any use of the middle field when in Create mode, and have to cram lyrics into 1/6th the space and constantly check to make sure one sentence isn't broken up into another sentence by mistake. Maybe even just being able to minimize one of three parts that are onscreen, giving more room for the other two sections. Hell, just being able to see the create field itseld would be wicked.
Love you guys at Suno, this is a great service, been having a super fun time for the last 2-3 months
Can you make an option to tab some famous name, for example, Cher or Garou, in "style".
ability to upload 2 song clips and have Suno analyze and create a version inspired by both songs.
Please please please fix the mobile site.
When you open the three dot "..." Panel on a song. The x is off the top of the android chrome browser, you can't click the x. So you have to back up and it reloads your page back to 1 so if you were messing with songs deep down the library you have to go back through the pages.
Speaking of that, the width of the menu where top bar navigation is, the left page right options to switch pages, you have to scroll it to the left slightly to get to that right arrow.
And lastly, when you first load a new page, and you go down the page and let's say you're going to click on a song to play well if you haven't selected that song already by long pressing, just the song itself, none of the links, just the body of the song , well just tapping play. Will move the page all the way back to the top and then you have to go back down since that just selected it and moved your page back to the top which often results and you accidentally clicking something and having to hit back. Which means you have to go all the way through all the pages again and it's just not okay. At all
It is horrendous.
Chrome / android
So there’s a thing I really hate about your interface on iPad. When viewing the library, if you accidentally touch a song while trying to scroll it pops the lyrics up filling a third of the screen, and this window can’t be gotten rid of, so you can’t see what page you are on or anything without moving the middle window back and forth underneath the lyrics window. Just put a close button somewhere on this stupid lyrics view please so you can see without having to reload the library and page back through hundreds of pages to get to where you were. Do I sound annoyed? Because I’m annoyed because this is annoying. Never have a pop up without a little x to close it.
It would be really useful if you could select the appropriate tags on the public version of your songs from the tags used to create the various pieces. For example, I had a rock ballad that I wanted to start with an instrumental orchestra piece, and now the published version of this rock song is listed as "soothing classical orchestral" when the bulk of the song should be listed as "indie pop, emo, electric guitar, bass & drums, male pop vocals, clear vocals, rock band with piano"
But, since it all extends from that classical piece, that's what it'll be listed as.
the audio quality is bad and you guys taking credit this bad audio in afrobeat genre or refund us credits when audio is very bad i reported half of audio i generated more than 600 credit bad expirence so far
The stems as midi files would be insane
Good day: I would like to suggest that the library could be sorted by date, and that there was partitioning by extension. Right now for example, all my extensions are clumped together, regardless of the song they extend, because of the original song title (i was lazy and different songs have the same name).
With V3 the tip was to add the world intro in the style prompt. Like "pop with intro"
It works pretty well. I am not sûre fir v3.5
It seems that full stems generation is difficult to implement by suno devs.
At least having two parts should be enougth in a first time. Instrumental and vocals.
And please dont use the solutions that are currently used in ai driven splitters available on the market. The separated instrumental track has a poor quality because some frequencies have been filtered out .
Allow for tipping of credits
Although this might lead to abuse on multi account users, so I would say it could only be paid credits
this, incidentally, would increase your $ per customer, and allow you to capture more out of your wealthier users (for whom this is trivial)
and they would be happy, as they could then support their fav artists
so kind of a win-win
would also increase engagement imo, similarly to how the contest has, but in a more organic manner, since it would be users who are effectively "paying out" so there's no algorithm to "game" it's more about finding an audience and giving them what they want.
If you include "tip" stats ie a leader board etc of tipped songs, this would at least put more pressure on botters/manufactured content by forcing them to basically pay a tax for unnatural views
so at least you make money off them
instead of them just using up compute and requests
another good but useless suggestion because will be implemented only in 2050 or never : add also the bpm in the song icon (where the song length is displayed) ***
It seems that you don't know yet the difference between the audio functions and the user interface function ... or you just pick up whatever you can to defend the dead UI programmer... the only one 🙂 RIP to the dead UI programmer. May his programming abilities rest in peace in the other world as they rested on earth the whole time....
great idea!
Furrybeans, I complain all the time only about the primitive user interface,, and you know that... and there are different programmers for UI and audio.. I know you must know this difference...
What are you actually asking be done to change the interface? Or do you just want it to periodically change for the sake of changing it?
Of course. But the audio upload feature required UI changes.
read my former postings (discard ist the wrong plattform for these things anyway, who can track a huge list???) ...as well as other ones postings.. there is no single change in the ui interface and missing basic features or bugs in weeks .. not a single one. .. so I assume their only UI programmer must be dead...
I actually did go back to the comments that Furrybeans was responding to before I asked. You didn't say a single thing about what to change, just claimed nothing had been implemented for weeks despite things being implemented literally every single week since 5/24. The only part of your rantings that mention UI were statements about assuming a programmer died. You didn't asked for anything at all.
Having had to click through even just 80 pages of songs to find something recently, the lack of a direct way to go to a page is a bit annoying so things like this are good suggestions. The comments about the developers are not helpful though.
That's why I was asking what you were suggesting.
Dear Furrybeans, who gathers the suggestions from here for the SUNO dev team?
The devs read these themselves. They comment in this channel sometimes.
As above.
There was a UI issue the other day where uploaded audio couldn't be extended if the browser window was too small. A dev responded in the bugs channel and addressed it.
BASIC FEATURES ARE STILL MISSING IN THE USER INTERFACE !!!
After 2 months of suggestions and complaining we still need a direct page selection (right now is even a bug there, the "programmer" forgot to check if you click more "next page" button than the total number of pages you have!)
- We need a feature to be able to give a note/comment to a song so you know later what you were doing first time etc.
- We should be able to give something like 1-3 stars to a song, to be able to differentiate between the best versions. Only a like button is not enough!
- Add a direct access symbol for both wav and mp3 download, the share link and get whole song in that line where other symbols are, so we don't spend so much time clicking and looking in that slow menu idea. There is plenty of space there to put everything users use constantly.
- add a "next/previous page" button (or even a search function etc) also the the bottom of the window, otherwise we make miles of long mouse movements every day
- add a 1-3 stars individual rating for a song, the like button is not enough, so we can differentiate between the best versions of a song.
And also the possibility to display only 3 star songs, or only 2 stars songs, or all songs marked with stars etc. - add a template with user favorite meta tags or a changeable default template so we don't have to put manually meta tags to every song (tags like "[Instrumental intro], [instrumental solo] etc . Because who wants a song with tags as in the actual default mode, with no solo and intro ? I guess nobody !
So why don't use the computer instead of the fingers in the AI age ! The DOS age is gone! - expand the search function! It is amazing how limited the search function has been implemented, and this even only after user complains of the missing of such a basic function! Please expand to this:
- look only in the titles, look only in the lyrics or look in both. 2. search also for partial words like "amazi" instead of only "amazing". A misspelled title can be never found with the actual function! Who had this idea to look only for whole words ????
Give us sort options for the songs (latest, earliest, bpm). Give us the possibility to display unfinished songs only (part 2 songs).
give to possibility to reverse the display order in the library from top up to bottom up (in case we look for some older songs)
- changing lyrics after extending songs delivers wrong lyrics. There is no chance now to postedit the lyrics! Enable editing the lyrics after the song was created with other lyrics than the original.
- add bpm in the icon of the song (there where song length is displayed)m this is very important for musicians!
- give the possibility to manually reshape the width of the columns in the app like most apps do. With a full screen, names of the songs are left and the often used functions are right.
With a wide screen that makes a lot of head movement every day! It is not ergonomical at all ! - like in the splice app, make a local saving folder for the songs and give also the possibility to download all songs at once that has been made in a certain day. How much time we spend daily clicking to download every song manually!
- and most important change: please change the UI programmer!
Most of these changes can be done in a couple of hours (plus testing), and any normal programmer knows that. A server based software can be updated often without problems, even daily. It is not understandable at all why at Suno it takes months to change/add smothing BASIC in the user interface! This is a programmer talking!
--------------------------------------------------------- ***
where do you know that from? you work at Suno?
I know that because they literally come here and comment
They've said it before that they read this
show me a UI programmer!
I don't know what each of them do, just that members of the team do read this
Actually, I do know that Tony works on the model, that was said recently.
I dont think there is a UI programmer here, he either is dead or blind...because nothing from these many suggestion about missing basic functions - has been implemented in many weeks! NOT A SINGLE ONE ! Not to talk about bugs in the UI !!! SO IF A USER INTERFACE POGRAMMER EXISTS, HE SHOULD RESPOND TO THIS !
Or they work as a team and it takes time to redesign some things like rating songs. And they don't just do things without discussing it...
Well, immindragon, It seems that you work for suno though, otherwise there is no explanation for knowing so much about the developers and the team and defending incompetence and laziness in such a obvious way...or you have nothing better to do.... So please send that dead UI programmer post mortum to a basic level programming course about lists, sorting and other basic things.
"Knowing so much" meaning "has been in this Discord for a while" apparently
but again the same question for Furrybeans, which he did not respond yet: if apparently there is no one who gathers here the suggestions for the devs and considering that they might have no time to do this themselves (fact is that 99,90% or the suggestions, missing basic things and bugs has never been implemented or fixed, although most ones are small UI changes), why this discord forum?? Just to loose our time writing in vain??
The devs do read this channel (and the others).
They sometimes respond directly to things, as ImminDragon mentioned.
There is no proof at all they read. Or that they care! Or that there is a single UI programmer alive ! Not a single proof! It is merely impossible to ignore all that bugs and suggestions for weeks and weeks!
Which normal programmer forgets to place a search function for the songs in the first released version or forgets to place a direct access page in a list with hundreds of pages in the second release?
I want to see the user/developer who applauds such incompetence!
And who could trust such a "programmer" who does not even that, to read here suggestions and bugs ? NOT ME!
So who gives me an email address of a product manager of a project manager to write him all the basic missing things in the UI ! Because it is a pity to have such a genial composer in AI and such primitive user interface in this AGE!
Another request: Chromecast songs to your TV or stereo
What im missing the most with the new insert audio feature is that we could add it to the song in any spot we want, instead of just at the beginning. sometimes we find a cool sample that would fit as a drop in an EDM song but you cant make it work, unfortunately.
We are all humans here. The Discord Crew and even the Devs. Its nice to get constructive feedback. It would be even nicer when it was a little less hostile against the hard working humans behind web interfaces and usernames. Of course they read here. We all read here.
come on, constructive feedback is being ignored for MONTHS! This is the issue right now! You ALL ignore ALL FEEDBACK, ALL BUG REPORTS, ALL BASIC NEEDS; ALL SUGGESTIONS from the users! And am talking mostly about basic things that are missing in the User Interface (UI),
How do you dare to say I did not give constructive feedback ? Did you read my postings before! You did not! Not only me I gave suggestions and bug reports, but many others before me and after me, but all IN VAIN.
And I am not talking about fancy audio features, but about basic user interface functions others have from the start! If you say you all read here what we write to make your life and our life easier, why is nothing done in weeks and weeks ???? And this in a web based software! As a programmer and product manager in SW and HW development I know how much time requires to add some basic sort and listing functions or other basic things in the UI! This is not a matter of MONTHS but of HOURS !
We are also humans, so why do you torture us and steal our time by ignoring all our needs ? Over 10 milions users! And you DONT CARE AT ALL !
Even here in discord you steal our time because nothing has been done in weeks in the UI! NOTHING !
We can really crowdfund a GOOD UI programmer to add all those functions if you UI programmer is dead or too occupied doing ... NOTHING or.. working for AUDIO which it should be his occupation. It is impossible in a web based SW not to make a needed and bug fixing update every couple of days! And there is much to do, alone the bugs !
Again: I am talking about basic things in the UI interface and not audio related stuff!
When nothing is done, somebody has to rise the voice, otherwise it will not change anything !
I can gather and sort the problems for you and send it via email to someone in charge, a project manager, if you give me an email address!
Because leaving all this for "every one at Suno is reading" did not bring anything !
Apologies if the mention isn't appropriate, but is it possible for any volunteers from the community to assist with the development of the frontend? I checked online and it seems that there exist only fulltime job applications for US citizens (I'm a Bulgarian citizen, plus I have a fulltime job already). Are you perchance aware if development is more so "closed" rather than open to anything like what I described?
Also a nice thing to have on the website is the ability to manually edit lyrics - extending parts really jumbles them up
The Discord crew is not directly working at Suno. We have better internal contact to some of the Devs but have no real influence on things. We can point them to things they might have missed reading here, what we always do with things they dont already know. The Jobs at suno seem to be more targeted towards US citizens. You can always apply from the official page. But to be a US citizen would make it more easy.
Thank you for the feedback! Always looking to improve ❤️ if you’d like to list out areas of improvement and post them here that’d be great! Always looking to find what members care about to better navigate the trajectory.
"Always looking to improve " - nice slogan! But it is only a empty slogan! Because nothing has been improved after user complains! NOTHING !
It is the first time in discord for you to write such a politician-wise banality like " if you’d like to list out areas of improvement and post them here that’d be great! " ???
What the heck are we doing for weeks and months ???? Or do you make fun of us with such a comment ? Or you David don't read at all to give us such an advice as we would not write anything in discord ??? I dont understand such comments and excuses....it is getting funny slowly....
You're correct that it's becoming a bit amusing and for reasons you might not expect, but in any case, with all due respect I feel like you might have missed several UI updates that have taken place recently, especially for mobile users (Android user here). While there are quirks to fix it's not like the developers have grown roots on their seats and have no lives of their own to live. The features that are missing do make using parts of the website a bit more inconvenient (a bit more so on mobile), but not impossible.
For example, I wasn't able to see songs in playlists with descriptions that were too long before - this appears to have been fixed on mobile for Android users at least (Chrome)
UI updates roll out quite often. We recently also received the Radio feature for example. So it's not like it hasn't been touched for months. Besides, it was explained above that this Discord server is not directly affiliated with Suno, so getting information through to them isn't quick and easy.
Oh I must have misunderstood, when you said “I’d be glad to gather and sort the problems”, I thought that meant that you would like to provide feedback on areas of improvement. If interested, please do. It’s incredibly helpful when members help point out things they dislike or would like to see present.
It seems to be more of a personal dislike of the dev responsible for UI
David, I only would do that when I have access directly to a project manager from Suno. Anything else, specially doing something here in discord has been proved to be futile.
Okay, feel free to reach out to them at support@suno.com at your convenience. And thanks again for your feedback
You are a patient soul, @dire frost
im curious what the devs thoughts on implementing negative prompts and how possible that is with this model
anybody suggested video improvements? like visualizers with soundwaves, etc.? that would increase the sharing on SM instantly
Edit: Sorry, I see they meant that as part of the video generation process
This channel is moreso a feedback for Suno the company. Perhaps you should ask in #💬┃general-chat or #🗣️┃chill-chat for more visibility
I’d like to be able to edit the lyrics displayed, specifically on full songs made up with many extensions. I have songs where the lyrics it shows just don’t match up right with what’s in the wing and I’d like to fix that
Song upload is restricted to v3 5 model usage and unfortunetaly is beeing limited to the known bugs of V3.5 . V3.5 seems having beein overtrained on some styles and than gives very poor results for others ( compared to V3).
What styles were better in v3?
A single example : Reggae roots. There are plenty others.
Specific examples are probably very important for any developers wanting to look into this, just saying - the more info the better!
This is has been already reported i think.
Every model is different, and you only know how it works with "working" with it. The Devs know what they have "learned" it but not how it will respond. So they are learning with us.
I guess by my side that the behavior of the model will highly depend of the set of training data. The most popular styles (in term of usage on suno server) will certainely get better . I would not expect that the others get worse.
Not only that.. but which parameters they used (there are soooooooooooooooo many settings!), how long they took to make it learn, and the consistancy of people might come in play too. As it learns music, lyrics and genre tags. But they are provided my people, and people are not consistent, so that could also explain some inconsistencies.
now I have a really huge doubt about the fact that if I upload the music generated by the AI on Spotify they can throw it down
Right. Suno is a young company. Tracking regressions is the goal of the QA departement of a software dev company. So we expect that in a short/ middle term suno wil have one.
you know what whould be interesting if we had ai make our own music video for a song
There are AI Videos services out there: #💬┃general-chat message
Heck, Noisee was designed for making AI Videos for music specifically
Making sure the AI never EVER repeats a verse (for no apparent reason), unless specified.
Have someone already said collaborations yet?
I do enjoy some of Suno's ad-libs and hallucinations, I feel like it completes the song more often than not~ but I can see why they can get annoying
Maybe having a 'creativity' parameter, like Midjourney's stylize, could let us tell Suno if it can run wild or stick more strictly to the prompt
It's still there... you can see where it generated from, both on the create page listing and the details
ah... it doesn't show on extension of the uploaded clip
it does sound like a bug
sorry I missed that details in your report or if not, it should be mentioned.
If an auto-generated title matches a song you've previously generated, add v# next to it.
I currently don't really edit the titles because they don't really matter, but when I want to find a specific one I can't so I manually edit v1, v2, etc onto every piece to set them apart.
Maybe have it only check public versions you have so it doesn't count the many, many songs that are private because they're not good
id like it to be able to 'get whole song' but exclude original upload....
that was a feedback for Suno the company 😊
Ah, I see that interpretation now, sorry about that
no need to apologize, I know the suggestion is far-fledged, but given Suno lets you download the .mp4s, I think there's an opportunity for them to hit it bigger than they do on the social media
i wish there was a negative value you could add, like things you dont want in the song, every time i make a song thats either dance or electro or something similar, the lyrics ALWAYS contain the words "neon light" and something with "city". i do not want that but theres no way for me to make it not happen other than writing my own lyrics
Please add ability to create vocals that are not distorted. Its hit or miss right now. Thank you.
you can get around this by adding tags like soft male voice, or gentle solo vocals. things of that nature within the song style tab.
using words like clear solo or things like that seem to help with the choir style of vocals.
you might be able to work around this issue by implementing more precise cues to the generator. maybe specifying a time of day or a specific location may help with this issue. just an idea.
One more please. Please add ability to: download song part stems (with vocal separated from music at minimum) and MIDI song parts (chords at minimum). Thank you!
Make a suggestion channel in form of a forum. So people can upvote and downvote and it's easier to keep an overview
because this one is being used to chat way too much
you can do it now!
to be able to seperate a part from it chain, so if one has a part one like you can detatch it from its previous links to make a song from that part and onward.
SUGGESTION: This has probably already been mentioned at some point, as it could/would be a common gripe someone may want fixed or implemented. Can we get some kind of audio trimmer with the option to fade out (if wanted for a particular song)? Nearly every song I allow the AI to make with my own lyrics always wants to start going on some weird tangent riff after you can clearly tell the song should have ended, and it just sounds goofy left on site and not pulled off to edit offline.
I hate deleting songs because even the fails give me ideas of different remixes and ideas so my "list" is blowing out. I really need folders or at least moving to a playlist removes from main list.
A notes option would be good. At the moment I'm using a convoluted naming convention in lieu of a notes option.
and how would you suggest i use that?
Suggestion: Add the possibility of obtaining the instrumental version and voice only, from your own generations
maybe the style box? to be honest. i dont gen lyrics. i try to write them. was a suggestion based primarly with tags in mind. my oppologies if my thought in head. wasnt accurate. wow im wording real good tonight. jees
If you want it to generate lyrics and avoid things like neon, turn on custom mode and write instructions for the lyrics, then click the button to generate lyrics. It should use what you wrote to help prompt the lyrics. Just tell it to avoid those common words in addition to what the song should be about.
I know it is quite unlikely and probably won't happen, but I would like regional prices. I am from argentina and $10 feels like $100 here
I understand your pain.. but the Suno team has to pay their servers too. And those price stays the same for them. Even if you are from argentina 😦
i know, i won't push it. besides people would take advantage of that and use vpn
I like to refer to regional pricing as "Why doesn't everybody VPN into Turkey before subscribing on Twitch? It's like a fifth of the price!"
Would be great to search your own previous songs . I can't seem to find the say 20th song I created and the name it was given can't be found either .
Negative prompt so we can say what instruments to exclude
I hope that one day we can copy and trim parts of a song and attach those parts to another song.
Would we be able to get a function that unlinks extensions? Like something like "Make Unique" so when you run "Get whole song" it doesn't include everything it was previously linked to?
Use case for this would be for example, my sample I've uploaded and extended off of. Doing an extension, I'd rather not include the sample as the start of the track. If I could then make the extension sample a "unique" entity similarly to Get Whole Song's effect is on the final product, that would be useful.
Especially in terms of coming up with ideas off of ideas.
Get rid of censorship! Music is about expressions and censoring that is an injustice.
(Cant make explicit lyrics)
As fas as I know, Suno isn't a diffusion model, but it would be nice if we could get a "Denoising Strength" option which determines how closely the new extension should resemble previous outputs, with "0" meaning "Do nothing new" and "1" meaning "Everything must be new" (So I'd figure an option like 0.8-0.9 would be nice to preserve a song's structure and make the extension much more unique, as with Stable Diffusion)
Udio does have the "Remix" button which is literally the function I described, but I believe Udio is a diffusion model because it even has a seed setting.
Would be great if suno created a song you loved and you could change a few words in it and keep the same tune and singing style you already love
Allow for free users to use 5 seconds or 10 seconds of audio, that way subscription users could get better results for paying and free users can get a taste of what can be used
how to find out all [metatags] and all (ad libs) that can be used
the ability to upload a beat and ai writes lyrics on the topic we tell it and record a reference
there's also this..
https://sunoaiwiki.com/
Suno AI Wiki (Suno Wiki) provides beginner tutorials and tool resources, gathering advanced usage tips and prompt examples.
all from the conmunity of users
You can do all that in a DAW and even make Suno be non-ear-destroying - which is difficult but sometimes possible. Now, though, everything is tin-can HORRID, and unusable. Unsure what happened.
And everything is cheesy as hell in 3.5, can't make anything that doesn't sound like a pop-crossover with another genre. Wish we could get 3.0 Alpha back, it was by far the best version. How they manage to go backwards in prompt coherence and quality is mind boggling
I've been putting in weird prompts to generate weird dub music and noisestep. That works about 1 out of 10 times. I'm getting some neat sound effects and noisy samples, but then it breaks out into a pop song after it starts the main part, LMAO.
They let users train the model
Oh god....
User ratings are used to train the model what is good or bad. V3A was trained by the subscribers.
Trained by WHO - people who only ever heard music through a TOY TIN-CAN TELEPHONE, because THAT is what everything sounds like now, LMAO!
Makes sense, a lot of people are listening to music on their speakerphones these days in some parts of the world 🙂
Yup, I actually said that to my wife! Good call! KILLS my ears on studio-quality headphones, though. Was NOT like that a couple of months ago. I guess I'll move on to something else unless I use it for solely weird dub and noisestep.
They have asked people to thumb down clips with bad quality or issues that should be trained out, but not everybody knows those buttons do anything other than sort clips
Good luck bringing it back to Suno and sharing with community after editing it with a DAW
so I can just drop tracks and gauge opinions in here?
If you have tracks made using Suno, you can share them in the showcase section
Suggestion: The Ability to "Star" a song.
Yes there is a "like" and you can sort by that but if you want better feedback from us allow us to also "Star" a song, Why? Because I often use my "Liked" songs as a library of completed songs. AKA a "Final" edition of the song.
"Why not sort by completed songs for that?" you might ask. Well I've found that using Completed songs to "continue" a song from can be very handy but not a good way to sort because I tend to have anywhere from 2-3 "completed" songs in the creation of a song.
This leaves some sort of function like a "Star" or "Flag" to quickly see my library of the "final" version of the songs in my Library.
would like so when one create a playlist its on top when ure adding songs instead of have to scrool down often to the very bottom
Being able to edit lyrics after the song was generated to match the edits after using extend.
I was going to share a reels video on Instagram today and I decided to use Suno.
My reels video was 5 seconds long, so I said, "I guess I can do it, it makes nice sounds, but while making the music, he didn't get any information from me, he didn't ask me how many seconds I wanted it to be, I think if the social media content managers of this place are going to use it, we should ask how many seconds we want it to be and we should choose that number of seconds and have them do it for us."
Is there any way we can stop the Create page from scrolling down automatically when a clip is finished loading?
Sort of driving me nuts when i get several ideas at once and submit prompts back to back while listening to clips made further up the list on the Create page.
not everyone is able to pay another 10 dollars for a VPN per month to get a subscription in Turkey 😄
the extension of samples with recordings of voices are absolutely HORRIBLE... the voice is so distorted, it's just a use of credits.... is there no setting for the distortion?
Most likely this is a limitation of the current models. Vocals for uploaded audio extend to have similar quality to the ones that are normally generated by Suno.
There are free VPNs out there. They aren't going to be as good as paid versions, but if your only purpose is getting five times as many Twitch subs for the same money...
totally OFF-topic, so sorry for this in advance to mods: I don't know how to tell you this, but more than 50% of the countries on the planet have a GDP that's lower than 5000. That translates to roughly 200-250USD per MONTH as minimum NET wage. When you have to choose between getting food on the table or saving money for a subscription that's more than 4-5% of your net salary, you don't think twice 🙂 For Europeans, the US, Can, Aus, NewZ, Kor, and Jap, 10 dollars may be nothing, but for the rest of the world it really means something 🙂 Unfortunately, most of the companies are aiming for users from this specific segment, while the rest of the world can just "try to keep up"
So we can upload clips now to extend. Can we get something similar but just for music or the voice alone? Not extending it but using it as reference? it would be cool to be able to have ai use our voices when generating the vocals
(at the very least it might offer more consistent vocals?)
Doesn't it do this already? (Though it would be nice to be able to cut the original uploaded audio off afterwards.)
For example: <#💬┃general-chat message>
is it possible to get the stems for each instrument in the song we create? would be a cool feature to add for premium users if possible
Inpaint from previously uploaded songs, allowing users to copy and trim parts of a song and then attach those parts to create a new song.
a switch to use the old lyric videos where the text didnt move in custom mode
And this is why we need either
- separate buttons for subjective "i like/don't like this song" and objective "this is a good/bad quality generation"
- a tooltip or disclaimer explaining what the thumbs up/down buttons do (train the next model)
train it harder to enunciate final 't's 'felt' is not the same as 'fell'.
Also, [End]. means [End] not [keep going from previous generations until you've filled in the two minutes].
Don't know if anyone has already suggested, or if the option already exists - but it would be neat to have Folders in our Creator library
Add Leaderboards
/chirp-beta
This. We really need improvements on this
Have the option to set a time at which a song ends for playback in a fade-out. Like at 2:35 end song for playback with fade-out.
When you Upload an audio file, the image generation should happen after it is uploaded, feels weird to wait for an image to generate before I can press Upload
Oh, also would really like to be able to see the accounts of people that liked my songs and get notifications if someone does like one of my songs. I'd like to return the favor, listen to a few songs of theirs and give likes.
New Tracks by accounts i follow :
would be nice to have a new tab in the Library that just shows new most recent songs by people i follow
This must be a popular demand by now but please allow us somehow to guide Suno to pronounce the words through the audio upload .. this will solve the main issue and struggle with the vocals !!
Is there a way to make A.I generated characters for vr performances of your songs? Maybe you guys can work with the company “soapbox vr” to bring it to life. There’s gotta be a way to have a Ai artist on a vr stage , that would be so crazy . Especially in mixed reality and the artist is in your actual room
I know my suggestion is probably , way out there. But I think it would be a game changer in way we consume this type of content
For music producers it would be extremely interesting to be able to generate vocals only / studio vocals to use in their songs
Allows us to only download the Instrumental without the Voices.
@hybrid crane do you know what could be interesting
right now when making songs on suno if you make 2 songs and you like both of them but are generated seperately you cant do anyting about it beside making 2 songs so what if there was a fuse option added you could fuse 2 song togheter yes it would change but it would stil be an in between of what both songs sounded like i think that could be interesting also the posibility to add a remix option sometimes there is a song you like but if you generated a full song you cant do anything about it beside taking the time exemple 3:02 and saying 1:43 to remix that part but the part will be intirely different so i think a remix option would be grate
I I think it would be extremely good to have an option to generate vocals only and one more thing, if we could do theese funtions, [girl singing] [guy singing] [the music fades out] [the music fades in] [electric guitar starts playing]
to the suno team: I hope the noise gradient and color noise issues are fully resolved for 4.0
i wanna be able to export each track individually after making a song - so like vocals, guitar, drums etc
When we get other pay methods? GooglePay (someone mean over google wallet but not work), Klarna, Automatic bank debit (whats normal in Germany), PayPal... Creditcards sucks!
A gazillion people would love to use stems to have each part separated. I would even go further, ho many times do I think "Hey that sounds cool", but there is an issue with the vocals (which have too many effects and sound autotuned). I would love to just regenerate on a stem (track) basis and even have the capability to keep prior versions of each track.
It would be great to have a way to cut the end of a song without spending credits, I mean, tons of times I get a song that last 2-3 minutes with a proper ending, but afther the end suno keeps adding extra things until it fills the 4 minutes of audio. The only way to fix It we have right now (AFAIK) is to "extend" the song from few seconds before the point it should end and use the tag [end] in the lyrics text box, but It isn't an optimal solution, since suno tends to make the same mistake again and again and keep adding things after the end of the song; and when It finishes the song properly you usually get the credits back because the generation is too short. The other way is to download the song and edit It in an audio editor but that doesn't work if you want to make the song public since you can't reupload It to suno.
That and convert it into sheet music.
I’d love to upload a song that may just need a drum track or to trial adding synth.to a complete song etc.
Create by, and display BMP, Tempo. Also allow for stems and MIDI download. This will make it a serious tool for composing unique songs.
The ability to extract the beat from a song with lyrics. Sometimes I like the beat but don't want the lyrics.
Lossless audio download
genre adding in the themed generations discord channel, WHy there is no punk genre added ?
british punk groups was around 50-60's till the 80's and was a very heavy scene
I just created it:) https://discord.com/channels/1069381916492562582/1252010299108757586
thankyou @hybrid crane
an option to define the end of the song, I spent all my points to close the song but it continues singing without having given it lyrics
Wait, we can ask for genres?
V3.5 has some trouble with ending indeed. Here's a tip: If you switch it back to V3 it usually ends normally.
If it makes sense.. I am not making all channels for the weirdest niches 😉
thanks
I always finish the songs with the following structure:
[Outro Instrumental]
[Fade Out]
[End]
It usually works pretty well
Can we just get an option to loop playlists?
we should be able to stop the current track from generating any further with the press of a button. i have 2 straight minutes of silence and strange noises when the main track already ended. this would also speed up the work process since it forces us to wait longer before being able to download it
Make sure to thumb down that track
please add a option to allow us to put any instrument we want in our instead of just typing it into the prompt too many times i have created a song when it generates it doesn't have all the instruments i want in the song.
The ability to Make a full song without having to commit the original uploaded audio would be awesome, I tried to delete the original file out of my library to see if I could pull the full song with only part 2 & 3, but it fails every time.
Thanks for all the fun with this all, it's been awesome!
i think subscribers should get 50 free credits daily just like free users
#💡┃suggestions-web i would really love a “more like this” option to produce outputs similar to a style in an output. Once I find a sound I like, and the song sucks, messing with extending it various ways often makes it worse, I want a stable “redo” of an output in same exact style to get more shots at getting lyrics right etc,
#💡┃suggestions-web I’d love the ability to choose maximum song length too - 3.5 is awesome and I can’t believe we even have this tech at this level already, but it tends to ramble and I think restricting it to less time, and maybe even setting a metronome speed, would be awesome and might help keep it on task. I don’t want 4 minute songs 99% of the time
I actually totally agree with this too. Any level of subscription should get all the benefits of everything below it.
I used up my subscription credits on day 3 not realizing I wasn't going to get daily credits. Yikes.
Have the AI song generator continue the lyrics if we extend it and it originally created the lyrics.
I'd love a feature that allows us to cut out sections of a song, using time stamps or even a very simple waveform editor for visual representation.
I also dont really get the pricing of the credits. 10 dollars would get you a subscription and 2.5k credits. But if you want to just buy more that same 10 would only give you 1k so youre better off just waiting until your subscription expires and then renew it to get 2.5k again.
The credits from a top-up don't expire at the end of a month
There are, I think, four different kinds of credits. Suno really needs to do a better job of telling you what you have and how they work, because it's basically impossible to know at the moment. It's possible to have three kinds at the same time, if there aren't more kinds I'm unaware of.
Maybe you can have four, if you keep your free daily credits when you pay for a plan and get those credits, I don't know
Well ill just wait until my subscription expires then and renew it then
I would love to be able to download Stems
Like i want to support them but if i just have to wait to get double for the same amount of money then ill just wait
שלום.אהבה.שמחה.תשובה.שירין
It will be good to choose mp3 oder wav ! For spotify and Co. It will be better wav. 👍