#💡┃suggestions-web
1 messages · Page 25 of 1
Cool, I can't wait to see your project then. Please make an AI song maker and we will test it out.
who talked here about AI other than you ! I was talking about implementing an audio editor to cut the end of the songs !
I said, you can make an AI song maker and implement those "easy" features. Would love to test it out.
Because, everything seems to easy.
they neeed 3 months to fix the missing direct access to a page in the library 🙂
so including an audio editor it might take 6 months at least 🙂. They dont have enough programmers... that is the problem. And dont have a dedicated UI programmer...
As long as you're not behind the scenes, you don't know what is actually happening. What if they are in the process of changing the entire back end? What if they have more pressing issues? What if they have a staff issue?
I understand your point. But we have to give it a chance. A month is not enough. Think about all those new released "games" that are rushed out because of deadlines with no testing.
If it takes a year, then it takes a year, but I'd assume they would have more than just an audio editor to show. Maybe improved voices?
you mix here AI stuff with UI stuff ! Independent things, independent programmers... you can work parallel at both... it seems that you talk about smth else and even did not understand what I was talking
ABOUT.
UI and AI are strongly related lol. It's an AI song maker... if they add UI features, they must make sure that the back end works...
You're talking about a song editor. You think they just add a button "edit song" and it all works like magic???
You must have short term memory loss
with no experience at all I made in six months as a student in the 90s, an external spdif interface for the parallel port as a master work, first with an MCU, than in an Altera FPGA, I wrote also the windows driver and the Windows application and made the first external audio digital interface for a PC. And I even documented everything... and Suno cannot even fix some bugs and add some basic features in the ready app in 5 months. And they are a company! Come on !
Quite accomplished I see. Again, looking forward to your next application!
An AI song generator with all those awesome features.
Because seems to me like you're comparing your work with the work of Suno.
Adding features isn't as simple as it looks. There is always something somewhere else that breaks.
man, I am talking only about UI stuff... USER INTERFACE !
Instead of complaining, just enjoy the app as it is.
A fun song generator. It's cool, we can download the songs, generate stems, we can mix song styles, we can use our own lyrics, we can create music.
Extra features will be added. When? I don't know. When they will make sure that it will be well done.
This app must cost a fortune to run. All those songs generated stored in their database, all those lawsuits being thrown at them, all those features being requested, etc. But this guy knows it all, simply click a button to add a new feature. It's so "easy"
you might have no idea of developing if you talk like that... 🙂 I wonder where you work at.. microsoft ? 🙂
but we all know that Suno has no dedicated UI programmer yet... this is the problem...
Please stop this.
If you have suggestions to make, make them.
I made about 50, even graphics 🙂
People like him are not worth keeping.
New features have to go through user testing. Some of these features take a long time to go through this process.
there is no more to make... because they are not implemented
Send em to Udio, have peace of mind
Hire him, he seems to have it all figured out. Look at his "impressive" history.
True, they do have opportunities: https://suno.com/careers
he he, EVERY every time I complain about Suno laziness or lack of programmer you here offer me a job 🙂 Thanks ! I dont need a job.
Seems like you do, you have all this time to complain
smart guys don't have to work much, dear.
In that case, I suggest you find a job real quick
I think a local McDonalds might be hiring near your location, apply fast
Maybe you can update their user interface? You know where you order yourself on the screen?
It seems so easy, I expect these changes in the next month following your employment!
why are you insulting? no more arguments ?
I am giving you job ideas, I want your talents to be put to good use! I want to modify my Big Mac when I do a custom order, but there is no such feature. I need an expert UI programmer to be on their team. Who better than "dj" grabillion who makes it seem so easy?
No insults here.
even more insulting 🙂 "nice" behaviour....and it is being tolerated
Please explain what is insulting?
Because I fail to notice what hurts your feelings right now.
Cause what I think is insulting is underestimating the work it takes to implement those "easy" features.
Probably this comment could be seen as insulting (I read it as a "go work in McDonalds" kind of way) even if you did not mean it that way.
In any case, this is a suggestions channel so I'd ask that this discussion end here. The team are working on implementing things that have been suggested, some things are implemented and are in testing. Further suggestions are welcome.
Alright
To be fair the other users hobby seems to be to insult the devs
Very aware of previous comments that have been made and have responded to them at the time. Would just like to avoid things escalating.
so what happened to u guys? ou tarted p with making alot of changes listening to community, and you have this section where people spew inn good suggestions every day, but NOTHING happens? your fokus is some rediccolous competition? common!!
I had writing after [Verse] [Male Vocal] or [Verse - Male Vocal] to accurate
Make the lyrics stop rhyming everytime it's generated.
That works?
Also it would be nice if there was a function to add voice effects like if there was a part of the verse that's muffled, it would be [Verse - Muffled] or [muffled] that would be nice
And the ability to add sound effects would be great as well
Sorry @rare obsidian didn't know that was an actual at...
Suno, please drop the new features soon, especially the one that allows me to cut the beginning from my song... ...I have such a banger of a remix to release... please don't let me wait too long to release this epic epic track.... (naturally inpainting would also be appreciated, but this would make my day)
Also, love you guys! (you don't have to say it back, lol...)
Yes. I just made a song and to emulate the sounds of a car I had to literally write "HEHEHE" and it sounds like tires speeding up, it's called "Addicted for speed" if you wanna hear
I cant find your song-
I think the addition of more extension features would be good. For example, if you want to add an intro or a section before the start of the song, similar to the one in Udio. I believe it'll be a gamechanger for Suno.
I think they just need to look at AIVA and...take some uh...inspiration
To be able to actually add intro, verse, pre-chorus, chorus etc manually as additional layers of a structure would prevent repetition. To be able to select instruments we want in the song, main key, front instruments, background instruments, gain levels (which sound is most prominent and when) would make vocals stand out better. To have a way to gate frequencies or a limiter to remove upper and lower noise floors would help reduce the amount of vocals that are artifacted. Same if we have access to a reverb "sub-setting" for each instrument or vocals (and to make sure we can select vocal register and grit and gender.)
I mean that's a lot...but it has to happen to compete and if you can make it more intuitive and simpler than AIVA. Then Suno here will have an advantage
Don't know if this is allready known...
I would like to have a function that allows to correct the text in a prompt.
A Remix-Function to the acutal Song would be nice to get the nearly same "prompt" but with every correction you want to.
I'm trying to extend the Song about the 20th time to hit a nearly same prompt i did in first place... but it won't work...
The first Verse goes to fast,
The last part goes to long and gets cutted by suno
Just a Remixfunction for Songs
Do you mean correcting lyrics? Best advice for that is, find a spot to extend from after the generation completes a template of verse / chorus / etc... whatever it is you like and want to keep. Then, extend from there and add "Reprise" to the verse and chorus tags, like [Verse 1 Reprise], to indicate to the model that these are slightly different lyrics; or use [Verse 1B] or [Chorus 2] etc. Fiddle around with it to see what works. If you did not label anything with tags before, label it now and hope for the best. Or you can just print the lyrics just like you did the first time and change what you wanna change and hope for the best. This is not a guaranteed solution but it does work, sometimes very well. If the sound quality is bad, do it again until you get something you like.
You can spend a lot of credits this way, but that's the price of perfection until Suno adds the function you're requesting.
What you don't wanna do is extend from anywhere further than 2min from the beginning of the song, because you'll lose things you like since Suno only looks back 2 minutes (in v3.5). Hope that helps.
That doesn't Work
Add a feature to refund credits for extended parts that are between 10-30 seconds long generated within the last 5 minutes (adjusted to appropriate time)
This would help a lot with songs that the algorythm keeps ending even with lyrics when you try to extend it to be longer. I've noticed it tends to end the song with parts that are between 11-20 seconds long which won't get refunded. Which in turns makes me waste a lot of the credits from my subscription
A more understated and square design to interest them
* {
border-radius:2px !important
}
.text-base\/5 {
font-size: .81rem;
line-height: .7rem;
}
.pl-10 {
padding-left: 1rem;
}
.pl-9 {
padding-left: 1rem;
}
::-webkit-scrollbar {
width: 6px;
height: 2px;
}
::-webkit-scrollbar-button {
width: 4px;
height:4px;
background: white;
}
::-webkit-scrollbar-thumb {
background: white;
border-radius: 2px;
}
.buttonAnimate {
position: relative;
display: inline-block;
background-image: none;
background-size: cover;
background-position: 50% 50%;
transition: background 1s ease-in-out;
overflow: hidden;
background-color:#333333;
}
.buttonAnimate::before {
content: '';
position: absolute;
top: 0;
left: 0;
width: 100%;
height: 100%;
background-image: none;
background-size: cover;
background-repeat: repeat;
opacity: 0.5;
z-index: 1;
pointer-events: none;
}
Re: Extend feature:
How about multiple versions of it, optimized for different use cases (or different lengths)?
For example:
- 30 seconds: Close the song somehow.
- 1 min: Finish off the last lyrics, and then close off the song
- 2-4 min: Take the song in a new direction.
I think making separate models for those goals would benefit a lot of use cases... Since those are commonly occurring scenarios.
It's quite often that the song is almost finished at 4:00, and then you can't end it properly.
Also, if we're thinking of generating pieces, it might be good to have a:
- 30 seconds: Create opener for song.
I didn't see it listed in known suggestions, but when downloading a vocal stem, could you also include a .lrc (or similar) file?
please make it so we can use credits in 5 intervals so if you get stuck with 5 credits you can spend them without waiting for the next month for your credits to refill
please limit trending song per user, too many of the same users
Album concept and ability to number/order the song in that album. 🙂
Any word on fixing the bug where you cant switch to v3 when extending from a song?
When you "cover" your own original song. It is instead called "Alternative version".
Cover is cool n all but what we need is "Feat" feature
Will there be a way to somehow link original songs to their covers? Understandably in a compact way like in a dropdown menu/a separate page.
Covers do link back to originals
Cover is nice, but it would be great to be able to regenerate a song with new lyrics. I have written a song in English, and I wrote the same song in French, and the only thing I can do is extend the English version from after the first chorus and use the French lyrics, but the result generated is different than the original of course.
I wish I could generate the two versions with the same singer, melody and instrumental.
I have used pre-chorus, intros, and verses or etc, even pre-verses
Hello, why i cant change the "image" of my music now? hehehe i create a new song but i saw i cant change the image anymore.
Forget it i can change again... hehehe
Hello all,
I have some concerns about this new "Cover" feature.
As long as I can extend someone else's songs, I can then cover their song and steal or copy their style.
This is quite alarming as it means that the uniqueness of the individuals on this platform is no longer safe.
Please, please add a feature that allows us to toggle availability for extends on or **OFF. **
That way we can protect out own styles and uniqueness.
I sincerely hope this gets seen.
Suno, I truely truely truely need a super simple feature that allows me to say: start this song at X:XX, end this song at X:XX... I'm willing to type it, if I have to, but with this remix competition and the new cover feature, I NEED THIS SO BAD. People that have been here for a while, know you use lot of extends for amazing songs, and if you make a whole song and the sample you extended from is at the beginning, it ruins the song. I have a remix of the floss song I can't wait to share, but it remixes the VOX and starts at 1:14... I can't put it live until I have this... I really really hope this will arrive before the end of the submission deadline...
well said. there was no thought put into protecting artists before releasing this feature and it shows.
Praise
You sunk that nail into the wood with one swing of the hammer.
We need something like a notification center.
"Buddy wants to extend your song? Yes or no?"
Something like that would be so helpful. You can even have people able to choose to set it to yes by default if they want.
until this is addressed i know plenty of folks that will not feel comfortable participating in the suno.com fishbowl
please be paying attention
i like this service, i do
but c'mon guys get it together
Or no by default.
Yes, I just meant they can choose yes or no. But I think privacy, like with Apple, would help their community longterm
It it a bit mind blowing Suno is protecting the rights of artists not using Suno by blocking their content from being uploaded, but not of its own users.
yeah what the hell is that, right?
very telling
they dont have our interests at heart and it shows
gl with that
Yeah as somebody that often has songs with multiple thousands of credits spend trying to nail prompting and progression, the idea somebody can just extend and cover it makes me not want to put anything public anymore.
they made it so you can only cover your own songs, meaning stuff in your library so you cant just steal styles but didnt think about just pressing the extend button to get it in your library? 
been saying that for awhile. and yet they have enabled one click theft of our "owned products".
This is the core problem here
Yeah it takes the problem we talked about before and now have made it a feature
yup, even if its link only you can extend it and rip it off
but ya'know, who would think of these things?
cartainly not suno
or maybe they did and just dont care
either way, not a good look
mods, please add a channel "talking about new added features"
or they could just address the problem. that works too
mods i love you guys, you're in a tough spot.. i get it. not mad at y'all
i was already bummed that they could extend and swipe
I've dealt with so many people doing this, a few even trended for a few hours before months ago when MY version got filtered
I'm looking forward to trending being full of "remove the cover feature" tracks which are all covers of each other. Not that I'm giving people ideas 🙃
My uniqueness is at stake 
yup its complete bs
i will gladly open up one of my songs to CCL licensing and we can get 35 versions of it on trending lol
haha love it
35 versions of cat incoming
Gib raptor 
I offer myself sacrifice
this will be resurfacing daily until this is addressed
fyi
plus... people already sharing 4 songs - and entire playlists of their covers ... what fun
Yeah this needs to be addressed in some way
lets ask flosstradamus how safe he would feel as an artist on this platform given its current implementation, how about that?
Guys did you hear these 32 covers I just gen
all trending Stei when?
I'm not Italian enough
Yeah I'm setting my stuff to private and informing my buddies in the music industry about this so they can set theirs to private
bruh ... no one listened to the first version ... none of us are going to listen to the 50 cover versions!!!
why not just make all your gens public? LOL
(i'm kidding ofc ... i listen to everybody's stuff ... love you all!!!)
ZING!
Agreed!
🇹 🇭 🇮 🇸
Let's hope it gets seen 🙏
There should be an option for the creator of the song to decide if they want to allow people to extend their song or not 
This should have been there for a long time already
Or reusing the prompt. Especially when youre selling subs giving exclusive commercial rights.
Could even make that a paid privilige with it
a 20 usd plan would be also useful. Sometimes 10 USD is too little and 30 usd is too much. It is difficult to spend every month 10.000 credits... but 2500 is too little too
Give me your extra creds. I have 10k and spend like.. Well, more.
xD
I spent 30000 credits in the first month 🙂 I dont have enough now.. I paused one account and on the other one only 2500
but my suggestion was for Suno... since ppl make anyway 2 accounts when not needing 10.000 credits
Imagine spending all those credits and hours crafting a unique style,
then bob from down the road comes and copies your hard crafted style for 10 credits.
That is the reality now 😔
you call flosstradamus an artist? that "song" is not even music! Even striking a glas with the spoon is music comparing to that soundclash
bruh
because of the cover feature they 've just released ?
This is a suggestion channel.
Don't attack people please lol.. ik he's not here but idk I'm not a fan of doing that. I respect your opinion ❤️
boy, is not "attack". is about that sound gathering somebody might call a song or music, which is not
he is involved in a contest suno is running at the moment, does that not make him relevant to this conversation as an artist himself? music is subjective. you are free to like what you like nobody is forcing your hand.
i know that. but the guy who chose that trash for a contest has not a bit of music taste... not even a slice of a bit. I would not remix that trash even for 25 grands
a contest of lack of taste
Similar to my suggestion and I agree 😄
what a pity for Suno!
Why don't you remix it into something you think is good 🙂
you cant make gold from sh**t
@tepid trench It seems the weird dolphin has found another target for its bad mood
while the Suno AI has a genial good music taste in composing (because was fed with a lot of good music), their creators show exactly the opposite with all these contest: the total lack of good music taste.
But the more modern trash young ppl nowadays like is being fed into the AI now, the worse the compositions become too.
You can notice that difference also between 3.0 and 3.5. The voice melodies are worse in 3.5, common, trivial.
I hope there will be a command in the future like "good taste music" or "trash music" so at least we can choose later.
✅ Warning logged for grabby_35252. I couldn't DM them. || You may not like the kind of music that has is being used for the contest but please don't be disrespectful about it.
Creating a cover of a song extension does not provide the option to 'Get Whole Song' 😦
idk maybe it's just me, but 90% of time making a cover is as good as creating a new song, with nothing resembling original. only bpm maybe but I'm not sure
Not entirely true. Given enough gens you can get the identical style. Same beats. Same instruments. I tested it.
you just generated a cover of the extension. how should it get the whole song from that? the cover isnt connected to the original chain of parts.
It's a little inconvenient to have the option to cover an extension but not have the ability to combine the cover with the previous parts of the song. If it's not technically possible, then maybe it might be wise to remove the option to cover an extension. Just my two cents
in some cases its useful to just cover the extension, for example if you have extended an upload and dont want the upload covered too.
Ah I see. I don't provide a lot of uploads so that didn't occur to me
there should be an option for musicians to opt out of having suno train on their music, also if a musician or artist wants to opt in, they should be given partial copywrite and royalties for the music generated
It could provide a new chain in the same vibe and sound for a song thats similar but different. Allows you to keep and develop your sound
Or you know, extend somebody elses song and leech their sound :)
I just realized that what I said doesn't make sense, seeing that I have no reason to perform a cover on an extension since I can just create another extension. 😶
If I am working with instrumentals, I will definitely do that.
Please, PLEASE, for the love of all gods, the universe, whatever you devs believe in, let us choose the denoising ratio on COVERS, I literally got 10-15 songs sounding THE EXACT same as the original(denoise close to 0) regardless of tags, then I got songs garbled up taking only into account the prompt(denoise close to 1). Why not just let us pick or if you are going to have it automated, at least put it between something reasonable like 0.4 to 0.6?
Unless of course that's how it works already and I just got like, extremely bad rng, but I -really- think you should let us choose the denoising 'or' have it automated if people don't know what they are doing. I just think the option should be there for people wishing finer control.
I will admit though, having the option to do a low denoising pass over songs does seem to clean them up a bit of that weird AI 'gargle/raspiness' that happens sometimes and make the songs sound a bit smoother so that's another reason to allow us to have finer controls, I'd 100% use that on finished songs to clean them up!
As it is, I doubt I'll use it beyond the free creds, between the rng of getting a good generation you now have to multiply that with the %chance of getting a good denoise ratio and that's a big turn-off for me at least am afraid. But at least since you are using audio2audio generation we are one step closer to inpainting and that's a good thing overall.
It seems that Covers tends to choose whether to honor updated lyrics randomly, and when it does, it completely regenerates with a new cadence and melody, rather than keeping the same one
I think we could benefit from being able to choose where we want the greater weight to go, whether it be style, lyrics box, melody, etc.
My biggest use case for Covers is likely going to be redoing lyrics in a 'finished' song
maybe something came in a little awkward, or there was a vocal hiccup, but otherwise, I love the tune itself, and I want to fix it
Wish to see in general: Tiberious has a good idea there, too but I would absolutely ADORE being able to reuse seeds. I work through songs in a strange way probbaly but i like the sound of one but mess up the lyrics, then using the same seed would be greatly helpful.
Great. Yesterday asked for the coverfunction now aviable xD great
Did i say great?

please add remix option for any song
ok they now prioritizing covers! like we dont have millions of itext to image ai out there we can use. this is starting to be suspecious, have they stopped everything with actual music making*? they get millions of seriously awsome suggestions who could be implemented quite fast if they just do it? Yes i love Suno its awsome butto me it could be even more awsome with all the creative suggestions from us users but nothing happens? Covers and competitions? who the hell cares!! common Suno get your lazy butt out of the comfy chair and start doing what u did in the beginning!!!
I'd highly recommend reading up on what the covers feature does. It has nothing to do with images.
ok then im gonna do that...
It's not a question of album covers
ok i just got so mad now because i have been here from the start and i see people giving Suno the most perfect suggestions every single day...and all i see happening is time goes and competitions! Butt yes this was verry cool even though actually doing something with our suggestions would be even nicer.
Yeah, because May of this year was "the start"
what you talking about i had 1 month off not paying,, are you a dev or something? do i have to proove i have benn with suno from the start?
had to re register thats how its work
I'm not a dev, I was pointing out that you joined in May
i coild start explaining why it says that but i dont really wanna focus on that, since its really not important, i have been loving Suno and using suno from the start and i want to tell them that avtually doing something with the users suggestions would be about time. they have had the time, money and plenty of good and relatively easy to implement suggestions so im saying. Why have a suggestion part when not using it
There are more features being worked on. Suggestions just take a while to get implemented.
There are beta testers, so presumably there are features being tested
Is it possible to add Klarna pay method?
And ImminDragon if u really wanna nitpick.. Suno started December 20 2023. at that time i was using it and loving it maby not with this username but still, using it is the important part with the wording "from the start" yes i missunderstood the "Cover" and did not read the details today and it triggered a long building up of waiting and waiting for even a !little! part of our suggestions to show up in the program
so how long can it take to implement that u can clean up your songs lyrics who get stomped on top of each other because of
¤extending" and ends up in the final result with no possibility to edit it? how long and how many beta testers do u need to be able to use the same voice,sound from one song and make different/other song with same style,voice? wich has been suggested mu maaaaany users over the time. Annyhow i dont want to be a complainer but more of a "Common" things are going very slow in the development. If they implemented just 2% of all the awsome suggestions pouring inn everyday they would be stomping the competitors asses bigtime. So i wonder why dontthey? the ideas its served on a plate, all they need is delegate and get it done. some of the suggestions would maby take 10 minutes for a decent programer to do, so why is it taking so long?
Are you saying you want to edit the displayed lyrics?
all im saying it goes veeery slo9w 😛
Things always take longer than you might expect. What may seem like a quick thing to add likely involves a lot of work internally to make it happen. I doubt they want to offer anything less than they are able to.
i understand that things take time, and i want lay this to bed now, and sorry if i sound mad but im more like a little kid waiting for something awsome to be even better 😛
That's how I start to sound when talking about the wait for V4
Understandable. It'll get there in time and there are more good things coming.
maby im afraid its just like all theese cool things that has been popping up from time to time in history and when they get moneytized all just oes to...h..e lll
There's a term for that, https://en.wikipedia.org/wiki/Enshittification
Enshittification (alternately, crapification and platform decay) is a pattern in which online products and services decline in quality. Initially, vendors create high-quality offerings to attract users, then they degrade those offerings to better serve business customers, and finally degrade their services to users and business customers to maxi...
ooh i dont think Suno is like that or ever will be, im more thinking of the expansion of the company, more employees more people to have something to say about the processes and ending up with human / personal disagreements andlastly someone leaves or starts another competitive 😛 more like that i mean 😛
that impacts the quality and development and experience for the customer
but hey im off to make some cool music...sorry for the rant:p and thanks for the responses
That degradation process is typically the result of near or complete monopolization of a specific marketplace, so not a huge risk when there are a few decent competetors
got alot of lyrics to edit now hehe
Good afternoon. For the second month, I have problems logging into my personal account. Since you have canceled the ability to enter offices linked to Russian phone numbers. At the same time, the money for the subscription is debited. they promised to transfer my account by linking it to the mail, but they did nothing. writing off money for services that cannot be obtained is fraud.
Please consider changing the maximal length for uploaded songs.. 1 min is sometimes too little. Why this limitation ? 1,5 min would be better for example.
And why cant the Ai cut itself the song to max length. It is kind of a struggle and wasting time to cut first the songs to the max length.
hey devs, here is place for a small second audio player to compare A/B versions of the original and the covered song in case the COVER function is active.
and It would be good to reduce the vertical space between the elements in the left side column (where the CREATE is placed)..so all elements are higher, specially the CREATE button so you dont have to scroll down
THE MOST ANNOYING "BUG" is to find the active playing song in the list... it takes many seconds to spot that animated logo of the song playing... why dont you change color of the text of the active song or make a colored frame around it ? It takes at the most 2 hours to do that !
I concur. Maybe just drop a hue or invert the color for the song playing.
On the plus side, you can actually select the text prior to clicking the poorly placed edit button and the text will remain selected. But yea... the lack of padding is a frustrating oversight.
I mean, you can go from the middle or bottom up. It just triggers me a little bit. Call it OCD. 😂
My natural inclination it to top left and pull down in order to highlight.
Oh I get it! It's bitten me in the arse numerous times!
...which seems to be the theme of their web development. 
I tried to volunteer to help with the UI. Noone replied.
For free.
I mayka da artz mun
I'm gonna loose all hope if this gets ignored like the last thing that made half my favorite channels quit 🫡 Please hear us
Add the ability to just have a clean solo vocal, i.e. no backup vocals and no effects.
WHY
In modern music composition many musicians process vocals with various AI singers and it cannot be done if backup vocals or effects (reverb, delay) are used. Make them "dry".
Thank you!
- Start songs cleanly on the down beat, i.e. on 1 (one).
- Show the songs BPM.
WHY
Its VERY tedious trying to get a Suno song to align with the metronome and at a place that aligns with a DAWs timeline. Most Suno songs start on an off beat or worse an obscure place in time!
For clarity, if the tempo is say 120 BPM, simply have the downloaded .wav stems start on 1.
Can we have a flag option (in the trashcan view )to select ALL files from the current page to delete them permanently? it's annoying to delete them one by one
This. Case and point on why it should be removed immediatly.
It took me 5 whole clicks.
for the web interface, you can use windows selection modifiers (SHIFT and CTRL) to select multiple tracks at once. This is me clicking a few with CTRL held down... then scrolling to the bottom and clicking the last track with the SHIFT key held down:
This also works for moving multiple tracks TO the trash and for adding them to a playlist.
thank you for the hint
why was this not documented anywhere?
Also, Re: Covers: Currently, it seems to target 3:25 as a song length. Is there a particular reason it doesn't go to 4:00 like a standard Generation?
Suno, folders are the most important item for users to sort songs properly. A playlist will never do what a folder does ! Take a look at Spotify how they implemented that. Drag & drop is a fine solution.
Dear Suno,
I spend 2k credits on the song (about a day of work), a full day on social media videos, a full day pushing/marketing everyone to like and listen and it's now at the number of likes and listens that it should have showed up in trending... at least English trending... but it isn't (it's no in old trending either), 34 likes on the song currently. That is soooo much work. And my guess is that is has been filtered for no reason.
Here is the remix-song, nothing objectionable there:
https://suno.com/song/bdcf34c0-8994-4ae9-bc17-c406055cd753
Here is one of the video's:
https://www.youtube.com/watch?v=21xsI26uL34&lc=UgxBAiR1kqzpoV-jYeR4AaABAg
I really wish I could appeal the filter, I worked soooooo insanely hard on this, but if it doesn't make trending, it will never get the views and a large amount of likes. And all that work will have been wasted.
We need some feedback on being filtered and some way to appeal and I really need this to be overturned, because this is such a vibe killer...
best, (Still love you, just a little less)
Sebastiaan
hello.
keeping it short and sweet. A decent basic feature for songs on playlists. If we wish to apply to cover art for every song on that playlist so we won't have to apply one by one, give us this option.
best,
spock
Cover feature - would be great to upload a full song I've done (not a short clip) and then cover it.
Why delete every song in the trash one by one? empty trash would be nice
You should be able to clear page by page by using Shift-click
did not work
Suno: using keybords key (key shortcuts) for the buttons, windows and dialogs would save us a LOT of time! It is very time consuming and muscle energy vanishing to do everything with the mouse !
Just an example: when editing a song name I have to press 2 times the mouse button to do the job. Once for "submit" and once to close the dialog window. A keyboard "Enter" could do this at once !
And an "Escape" for canceling and closing the window. Same with the player buttons, next/previous song: add keyboards key to those functions ! Is not a big deal for a programmer, but a big deal for millions of user.
sorry. i am stupid. i understand how it works. thank you
It would save us a lot of time to reaccess the prompts we used to generate lyrics with chat GPT ! So make a FIFO or LIFO list of those prompts so we dont have to rewrite everything. Sometimes I only change a word for the next song.
hmm... silly question on that, but why not copy it into a text document?
Then you can take it between other ChatGPT models and have them polish it up further
Can we have a better time stamp system for the sample extension selection could it not be a var min max between 01 and given max time stamp? Right now its just choose between minimum and max selection should be easy to declare vars differenty in python well i assume its in python
Suno Alpha site leaks deleted data into Google
If I delete content my expectation is this shouldn't available, but Suno's alpha site is being indexed, so I can just find my deleted material on Google. Suno's production site is not showing this content in search which leads me to believe the two sites have slightly different search exclusion rules in place.
Suno Alpha Redirect Rules not Applied to Profiles
This might be a bug. I'm not an alpha user, so I'm not able to do much on the alpha site. This is expected. However, unlike other parts of the site, the profile pages have no feature flag check in place, which allows me to navigate directly to profiles on the alpha site. This seems like unintended behavior given the redirects in place elsewhere, so this might be something worth patching.
COMMENTS! 😭 please add comments
I do this already, cause no alternative... but I should I spend time switching windows and using 3rd party tools if a programmer can simplify this in 20 minutes for 10 mil. users ???
Suggestion: cover feature also available for playlists. It's okay if the number of songs multiplies the credit cost.
I have not been able to get downloads to work in any format either through the browser on my phone (or the iOS app)
@clever harness what is "Suno Alpha site " ??
It looks like a staging site prior to rolling out changes to the main domain. It's not hard to find, but I don't want to broadcast it in case the Suno team would prefer that not be widely known. 🙂
I see, thanks
Allow us to make covers of stems
I have gone through almsot 100 of the free covers and it won't take the new lyrics.
Please don't autoscroll songs list on timeout. Like, you scroll the library to find the right track, start listening to this and that, and out of nowhere the page jumps back to the bottom of the list. Irritates as hell.
Add [Unheard] Filter toggle (next to “[Liked] | [Public]” buttons ) to filter only songs with the Red Dot
Now that we have covers we need longer audio uploads even more.
So yeah, please allow longer uploads
I would love to actually be able to rip the vocals out of a generated song and have a clean instrumental that I personally can sing over
Well, at current state, I'd say covers are not functional. Out of every 10 gens you make, 3 are exact copies of original, 6 have nothing common with original, and 1 is actually an alternative version which you probably won't like, so pls make 10 more gens. I'm sure it will get better eventually, but yeah, propper stems and seeds would work so much better.
/chirp
Chirp is retired, and no longer on discord. You can make music now on our website: https://suno.com
FARGO long time no see. 🤘
Suggestion: On a public song page, link to all available covers of that song.
SUNO, the 60 seconds limitation for the uploaded songs is a contradiction to the cover function! Cause 60 sec is not enough if you want to cover a song! You cover only a third of the song !
So please change that to 3,5 minutes... or at least 2 minutes...
You guys are useless
✅ Warning logged for spare.dat. I couldn't DM them. || Don't be disrespectful.
I didnt succeeded using cover with an uploaded song.
Yeah ! A big credits consumer probably
do u guys know a tool for only get the beat from a suno song
So far using cover feature, the first half goes well but once it gets to 2nd half, the vocal no longer follow the original song's and just falls apart~ Gonna wait for updates and try again in the future~ 🤗
bump
and also limit trending playlist... same issue
Yeah I'll throw there, and delete this one. I'm also recording a video of it, so it's clear to see.
Longer uploads could help to give it more context - For both extending and covers
That's why it's needed
Chirp
Hey yall. Dont know if this will be seen by the right person, but i certainly hope so. When it comes to the AI, compared to anything else on the market that is publicly available, this absolutely takes the cake. HOWEVER, with that said, I have found the AI doesnt understand specific genres. Most notably, Riddim. Yes i know, not everyone is a fan, however I am an artist and use Suno A LOT for ideas and workflow when I am making Dubstep/Riddim songs. I would suggest the AI learn what Riddim actually is. I don't know if this could be done by simply having the AI "listen" to or evaluate popular riddim artists, but if so that would probably fix the issue quite fast. Examples would be artists such as Wodd, OG Nixin, Artix!, Izzy Vadim, Virtual Riot, HOL!, Crankdat, Jkyl & Hyde, Monxx, Point.Blank, etc. I imagine with the AI model there is a certain kind of "learning" mechanic so that it can adapt and better understand user inputs and requests. I sincerely hope to see the input of "Riddim" or "Riddim Dubstep" in the song description box closer match what Riddim is, especially with how popular Riddim has recently become with HUGE EDM artists transitioning into the genre.
I’ve been using cover to redo v2 songs which originally had bad sound. So far I’m out of my free credits and I have one usable cover. The issue for me isn’t that it’s a duplicate of the original (I wanted that, and it mostly does a beautiful job of essentially remastering the lower quality v2 gen into a good sound) but that it only “covers” the beginning of the song and then it loses the plot. These are not long songs, they average 2-3 minutes, and after the first couple verses and the chorus it does whatever it wants. In particular it doesn’t bother to keep any good instrumentals and nice finale in the original but just plods on to a boring end after a promising start. If I could do the beginning and then say, “Okay now do the second half the same exact way,” I would have something.
Hey Suno, the cover feature is simply amazing!! I’ve been waiting for this capability for so long. It’s such a delight!
I have a request, and I really hope you'll consider it. If I understand correctly, the idea behind covers works similarly to ControlNet, but for music. It can extract specific characteristics from the original file, like the vocals and melody, and use them as a reference for creation. I believe this could be incredibly powerful for creators if a similar feature were available when pressing "extend" in Suno—allowing users to provide a reference for the extension that guides how they want it to sound, including melody, vocals, etc. This could be a really strong creative control tool.
Thank you in advance! I love what you're doing!
By the way, out of all the amazing AI innovations released this week (and there have been some great ones), I'm enjoying experimenting with Suno the most. So again, thank you!
If such a feature becomes available, it would also solve the main challenge with covers, which is creating a full song all at once. With the ability to apply covers to the "extend" feature, people would have much more control and the ability to fine-tune the complete outputs.
you should work out the bugs before you start charging monthly fees. where'd my song go still waiting for an explanation
wish I could be more helpful with the feedback but that's all I have, I'm a bit baffled because it's been working fine and right when I finally created something I really liked it magically had an error and disappeared. I'd like to think you can understand my skepticism at this point
Piece of Cake 🍰 And REALLY needed...
Give us a LOCK so we don't accidentally delete a song or extension (again) while clearing out 10 gen fails for the 10th time. Please. Can't delete without unlocking first. that would be SWEET 🧁
IDK if this was ever requested either, but a feature to make fandubs/translations would be very neat. Where you can put translated lyrics in and it can resing an existing audio with the translated lyrics. It would make it easier to create fandubs.
Possibly the main issue with this would be that you'd need to have rights to the original audio as Suno generally don't want users to be uploading/inputting content that they don't hold rights to.
Ahh yeah, I could see that being an issue 😭
I have made original songs i’d like to have translated so I feel like it would work for that as well. Idk, I’m a big geek for dubbing and have been experimenting with this site and another for making some fandubs
it seems like all tools struggle with suno songs to remove vocals...
Begging you with all my heart to enable the "extended from" drop-down for "FULL SONG" versions so when I need to find the original generation I can do it without spending an hour sorting through the ridiculously unwieldy library.
Or, and this would be better, just let me have a basic browser with creation date sorting capability. The search is not useful.
Even if I could search a title in the current library and then filter by "liked" it would be better, but that is not currently possible either. On Mobile or desktop. Unless I'm missing something.
Theoretically Covers can do this. I've not been successful yet with instrumental but I have gotten vocal-only results that are far better than the stem ripped versions, using prompts like "solo vocalist, sans accompaniment, vocal stem, remaster".
Not reliable at all, but it has worked once or twice.
Cover work great for me, even some songs are just copies and some very different, the OUTPUT is still amazing ! I got amazing versions of almost whatever. Only one song did not want to be covered at all. And it eliminates also bad audio quality of the songs if you get an identical one.
Cover is an amazing cool function! Only the limitation of 60 sec uploads has to be removed to make it perfect.
I 100% agree. A lot of the time I take a suno song and bring it into the DAW for final adjustments... trimming endings etc... But now I can't really do a cover of the final version since I can't reupload the final version which is over 60 seconds. Seems like such a pointless limitation.
It would be great to have some options for cover, to force the AI to do more certain things like: same song version, keep the voice melody, keep the drums, keep the guitar, keep the same vocalist and specially: REMOVE voice and make an instrumental version.
wow, what a good idea! To get only the voice from a song it is a dream of many ! Cause you can do the instrumentals in the daw with better quality, but such voice you dont get easily !
Here is a work around. Use stems to make the instrumental version, and cover that one 🙂
oh 3Daizy, this is a genial workaround !!!! I kiss your hand for this hint !
Why can I not do a cover of a stem? I assumed I could cover the instrumental stem and basically use that to get a higher quality instrumental
ah, it does not really work in one step for what I thought it would work (to get only the vocals) cause with stems you have the 60 sec limit ! But there is a workaround also for this limitation: you cut the vocal part stems in 2-3 pieces and cover them separately (in the hope the AI will sing similar) and after that you glue them in the DAW so you get this way a 3 minute vocal track... but better would be if suno removes the 60 sec limit
I agree 🙂
I am gonna try this tommorow... now I am busy covering. But maybe someone here can try to see if the AI sings the same with 2 different parts of the vocal only stem
I have noticed the AI understands perfectly any instrument of the song it covers and any vocal part and even the lyrics.. so it is really intelligent 🙂 like a real musician. That is why it can be told to keep some instruments from your original and he would know what to do
another cool thing that could be done is to imitate a drum or any other instrument with your vocal, uploat the file and tell the AI to play this instrument in a similar or identical way as a separate, single track - not as a complete song !- and with cover you could make variations of this. Did anybody tried this with the upload function? maybe a prompt "only drums please would work".
I haven't tried it with cover but it worked when I imitated a trumpet with my mouth and extended it into a song, so I expect it might work for cover in a similar way
but probably the AI is not used/trained to make tracks with only one instrument, only songs. So there must be implemented a special prompt for this, which also should be obeyed. And besides that, he could recognize that you sing and try to make a vocal track
60 second limit just needs to be removed. Have any devs commented on it?
It doesn't give options to do cover of a stem
No but you can upload it. I didn't say it was perfect (yet) 😄
you can't upload more than 60 seconds so again this wont work for a full song
It just takes persistence, experimentation, and more persistence.
You need to cut a part. Like Grab said: #💡┃suggestions-web message
yeah but then you lose the other parts you wanna keep
I have no answers beyond. I am just a moderator. So hopefully devs will find a way to extend the upload
It can work. Just clip your song into overlapping 60s sections, like 0:00-1:00, 0:50-1:50, 1:40-2:40, etc. Then keep generating until you get something you like that fits with the others. It's not easy, but it's quite possible (I've done it).
and then you stitch it all back together in DAW after?
Affirmative. <#📚┃suno-school message>
As with everything, persistence is the key.
Hmm yeah thats super inconvenient vs them just increasing the 60 sec limit
Nice.. you should ad it as a https://discord.com/channels/1069381916492562582/1252267444773326848 too!
I'm just offering you a way to accomplish it now, if you want.
yeah i appreciate it
Once I'm done experimenting, for sure. But it's not a workflow so much as a desk full of papers and banjos at the moment.
has anyone tried using it to fix instrumentals from stems?
Same answer as above.
Just listen to the two versions of "Fenchurch" in my post to hear what is possible even as a first attempt While I'm experimenting.
It'll make you say holy sh&t.
It's always easier to have one click solutions, but out here on the raggedy edge we gotta make our own ways to be happy.
Devs just need to remove 60 second limit and also allow covers on stems.
so guys, here is your dream coming true with a workaround: separate streams of relative good quality!
fact is: the AI could easily make also instrument parts from a suno song or an uploaded audio. It only has to exist an option for it to obey this.
If suno implement this, we can make separate instrumental tracks (stems) from any song!
- So you take a good song, you make one cover which should become identical (keep trying until you get it), this gives a better sound quality.
- You separate the voice and instruments out (better with external program/site eventually
- Upload the instrumental 60 sec part
- Make a cover of the instrumental or the voice so it becomes clean (keep trying until you become what you want)
- Make a cover of this cover telling the AI only to keep the bass, drums, guitar of what you want (is this command is available).
- Keep going with the other instruments untill you have them all.
7-x. Cut the next 60 seconds and feed it to suno. Repeat the above. If Suno removes the 60 limit, then these steps are not needed anymore.
x-y. Glue everything in your DAW
They though could teach Suno to make the above alone, several times in a row so he delivers alone the single instruments tracks from a given instrumental song which already exist!
Another suggestion for better visibility: change the color of the "full song" logo so it is different than "Part x"
Overlap the clips by at least 10 seconds to save yourself some editing headaches.
We're calling this method a Covremaster by the way. it'll catch on.
btw make covers free.
oh nevermind its in beta I have adhd sorry.
I agree with you DJ Grabillion!
I always hate that 60 second limit.
✅ Warning logged for spare.dat. I couldn't DM them. || Don't be disrespectful.
Ok.
weird, I thought the AI listen more to the lyrics than to the audio stuff in the cover mode. It seems it is the other way.
By random, I made a song were an important word at the end of the verse was left out by the AI, but I did not notice somehow. And at the end I made covers to get a better audio version. But the AI is still leaving out this missing word in most covers I created (at least 20 songs). Only in 2 covers the AI is following the lyrics.
So dear Suno devs, train the model to listen more to the lyrics than to the audio, cause this way we will spend a lot of credits to fix a small problem like that which should be fixed with 10-20 credits.
This almost always indicates an issue with your lyrics structure. Look for invisible characters that might've snuck their way in during cut and paste (just delete the entire stanza and retype it manually).
Remove tags and special characters like ~ or ... from the vicinity of the lyrics.
there is no such character... even with another voice and style he left out the word. Only 2 versions from 30 has the word... so he can do it theoretically.. but he avoids it
ha ha, now he made a donald duck voice ! First time I hear that after thousands of songs... it is hallucinating !
And now it makes another weird effect: like slowing down the speed... it never happened to me so far that a word is left out... in this song even 2 words at different places. An never 2 such weird effects one after the other ! It is crazy. But I am sure it is a temporary problem caused by an alien attack or smth similar 🙂 Not a Suno problem actually. So I have to wait till tomorrow to continue until the alien is gone 🙂
A little feedback that I think is important (and for stuff like this I'm not sure if you would prefer if I talk to a dev, I only actually have contact with one of y'all lol) is that I think you guys may run into legal issues without the cfg slider, even if it is confusing initially for some users. It won't be for a while though, but when people say "oh, it's gambling," I think there's no real basis for this in normal gen AI as it's an intrinsic element of AI generation, and there isn't really much you could introduce to help control the output in a manner that would substantially reduce the randomness.
However, it seems clear to me that the "two gen" system generates with either two sets of arbitrary parameters, or an LLM fine tune is setting those parameters, but in the case of audio2audio, it really does mean that some element of control is obfuscated, one which would substantially reduce compute if you know how to use it.
More importantly, at times you may be looking for a high denoise (as I just was and spent a lot of creds on) and in the current system, that means rerolling until you get it (obvi it could be the denoise is fixed and the seed needed to break out of a set pattern is what I'm rolling for but really it's irrelevant to the argument, and I strongly suspect the denoise is not fixed anyway).
The issue with this is you get into the loot crate territory with this imo, where they lost that lawsuit pretty badly in the EU. You guys are sitting on a product where you have the stochastic dopamine that you need to make it entertaining and addictive, but much like a normal game or creative activity is: however, if you intentionally introduce additional elements to boost this, because there is precedence set and y'all are already under attack, I would presume someone will open a class action down the line, even if it wasn't intentional.
Of course, I'm not a lawyer, and I could have a total misunderstanding of the back end as I'm literally guessing.
Also, I mean, I would appreciate it personally as it would allow me much better control and understanding over my outputs, and I could better understand what token is affecting what, whereas right now it is tough to tell at times if a particular token has a certain impact, since it could just be denoise or cfg was shifting.
Even if there was an automatic v. Manual transmission type thing, it would be useful
I don't like posting stuff like this in pub discord personally as I'm sure I'll see it popping up on reddit bots in a couple weeks lol, but I don't want to dump on the ML departments inboxes when I'm not sure if this is their wheelhouse
And I'm not just gonna dm the senior dev guy bc that's improper, so again, apologies if this is inappropriate to post here.
We desperately need longer uploads for covers!
Will be a great feature to regenerate song in better quality
Would love if all thumbs-down tracks were sent to Trash by default (maybe add a toggle for this in Trash?).
you better delete your post! What do you want? To scare them??? So they remove or change the good feature ? 🙂
No, to add in a new feature actually
I'm saying that it is missing a particular feature that Im sure is implemented on the backend
And that it's absence could be problematic from a behavioral perspective
Ie highly addicting
In a way that could be argued
we can see from Udio that you can force the AI to a big extent to do certain things now are random in Suno
Well, we actually don't know that they're random. I suspect they aren't just bc Suno is AI maximalist
Like, they use LLMs like they candy
Imo
so it's prob a fine tune to improve user experience
The issue is that this still could cause issues with addictive behavior, and at least having the ability to use "guidance assist" vs manual
Would make it much less of an issue, while also in a technical sense improving the quality of the overall product (although, this is more wishy washy as the added complexity could confuse new users, but again, just make it default to assist)
It also may not matter, frankly, as I am not a lawyer, just a guy whose at the right nexus to observe this, namely I'm an ex junkie and also love ML
Friends and I all encounter a problem, when the lyrics are in Chinese, the vocal will have Cantonese (? seems to me ) in first few word, then switch to Mandarin. Chinese has traditional and simplified but they are overlapped. The Hongkong people speak Cantonese (Traditional Chinese) , the Mainland China use mandarin. (Simplified Chinese) This bug often let people can't understand the lyrics.
I'd really like control over how different the two generated songs are likely to be from each other, and for covers how close they may be to the original song. Having reasonable defaults is OK but it is wasteful all round (credits, time, compute) to not have control over this when needed.
This would also be useful if more, but shorter, generations were offered (as suggested in one of the recent polls) because then a range of songs with more narrow or wide variations could be generated so you'd get a nice "palette" of generations to choose from.
The utility of covers and extensions would increase, along with the expected "inpainting" function.
Downside to this? I guess additional controls could complicate things for users, particularly when you're trying to control AI which is very random in nature anyway as the effect of the controls may vary between generations. I don't think it'd necessarily be difficult to understand what such a control would do if documented accurately. Even then, after trying it a few times it should become clear what effect it has.
I think more but shorter would be so frustrating, id barely use it tbh. It's what kept me off udio until recently. They just need to give us denoise access and/or cfg
Would fix it 100%. It's still a great feature, but yea, sometimes I'm looking to clean up or polish
And sometimes I'm looking to see where this start point can take me
But also, with granular control, I can get there faster
And find places I wouldn't
I mean as an option. e.g. 2x 4 minutes or 4x 2 minutes or 8x 1 minute each time you generate, that kind of deal (as I rarely want 4 minutes in one shot)
Oh yea, that would be cool for sure
But it would have to pick a section from the song
So they'll need a UI update
But that could be nice
I'll never say no to more features, although I do understand that new users need a streamlined experience
I wasn't aware of the switch in mid-stream, but I've heard many say the songs marked "Chinese" (not Mandarin) default to Cantonese. It may be a preferred language of Suno when defining Chinese. Have you tried "Mandarin" and gotten the same results in Cantonese?
I wrote a song about Wuhan, but I used chatgpt to interpret it. I hope it's Madarin. Lol
Please allow "full songs" to include link to the original song if it's a cover. Currently, only extensions show link to original song and where it's extended from.
Accidentally deleted a song trying to remove it from a playlist, then it took a long time to find in the TRASH because there is no search.
Please add a search feature to the Trash. And, if you have the link to the track, add a RESTORE TO LIBRARY contextual menu option if it's in the trash as another way to avoid this frustrating problem.
I was thinking a lot about my workflow with Suno and I suggest that there should be an option for pro users to have a notion of “project” instead of a plain infinite list of songs. I think it may be more convenient for people to work on a specific composition in a kind of sandbox , so it’s easier to not get lost in 100s of generations, extends etc.
There is a Restore to Library in the menu of the song when looking at the list of deleted songs. Is that helpful?
That works only if you can find the track. My issue was clicking through the 16 or so pages of trashed items and I couldn't find it. I ended up having to use the browser Find on each page because my eyes couldn't identify it in the list.
It was the most recently deleted item, but I could only guess that it was sorted by creation date.
The workaround would be to create a private Playlist per song you're working on. Every gen, add to the playlist. When done, you can thumbsdown what you want ..... hmmmmm does that remove them from Playlist? I don't know. But , a mass move to trash of your thumbsdowned extensions should be easier then too.
You might not be tracking the deletion date to be able to sort by most recently deleted, but that could be another way to help.
So that works specific to your request for ".....add a RESTORE TO LIBRARY contextual menu option if it's in the trash....." statement? Great !
The deletion itself happened because I accidentally added two copies of the song to a new playlist, and then when trying to remove one of those from the playlist I messed up and chose Delete instead of Remove from Playlist
Yea. It's a bummer. Been there too much. Hope they put a search field for the trash as you suggested. For me it's my fat thumbs on a small mobile screen that does me in.
I had the link still to the song, so if Restore to Library could be added here also (if it's in the trash), that would have be cool too.
My crisis has been averted for now at least.
Workaround: Maybe if you add to a Playlist, then it will be in a "list" again contextually, then scroll down click the [ . . . ] and choose Restore.
I tried that actually.
It seems you can still add a trashed song to a Playlist? Idk
If the song is "in a playlist" but in the trash, it won't actually appear in the playlist.
Oh ... darn it. Ok. Worth a shot.
on Mobile App - Lyrics:
ON/OFF toggle switch to “Autoscroll”
(ie: Spotify’s scrolling lyrics function)
i dont want to trash any songs, i can always chop samples from any outputs. sometimes you miss some gems on some tracks buried in the middle or end and only heard one time on a rushed play through and if you trash them they could be lost forever : (
I think the same, I never delete songs... some gems hidden in many songs which are not perfect
It would be great to have folders and drag & drop some time...then we could sort everthing
Is this the only place to post suggestions? How would SUNO see the suggestion? 15 seconds later it scrolls and is gone.
They pull all data everyday and filter out the normal chat.
funny, they implemented one of my very specific suggestions
I have a suggestion (probably been made before). Either before or after you create your song, you should be able to put how long you want the song to be. Would probably work better if you did it AFTER creating the song as a sort of ending edit.
pretty difficult to calculate... so many parameters: bpm, lyrics, song structure, intro, solo etc... not really wise to start with that
Yeah. thats why it would work better if you can edit the ending after creation. I have recently learned (today) that putting [end] doesn't do SH*T with ending the song. It will create the song as long as it wants. 😆 BTW: That is a glitch. If the song has lyrics, it should end the song when the lyrics end, but Nooooooo, it ends it when it wants. 😆
with the cover function I saw for the first time a fading out ending from itself, without a prompt .
I don't delete things either. I just use the trash as a place to store stuff I don't want to see anymore. It's not like it empties itself or anything.
Some features that i would love to be added to Suno in the future at some point (sorry in advance if any or all of these have been suggested before)
- the ability to delete multiple songs at once by ticking a box like with Udio
- the ability to remake/edit a song with the same voice, melody etc in case of misspellings, wanting to edit certain parts of the song etc
- the ability to use the same voice (or voices if Suno makes more voices in one song) in our next songs
- the ability to delete more songs at once in the trash (same as the first suggestion basically)
- a search bar in trash to find accidentally deleted creations easier when we (like in my case) have 100+ pages in the trash.
You can select multiple songs at once by either control-clicking them individually or shift-clicking to select everything in a line, like in Windows Explorer, then just select to delete
I might be wrong here, but wouldn't the cover feature deal with number 2?
Oh, i didn't know about the control and/or shift clicking thing, and the newly added cover feature barely works for me, idk if i am doing anything wrong or if i'm just unlucky
The ability to select multiple is not well known. I try to let people know if I see, just because I don't think almost anybody knows about it.
Ah okay ^^ thanks again 🙂
I don't really know much about the cover feature, since I'm on a free account and can't try it yet. It will probably be the solution, once it's working.
Browser Suggestion: playlist drag and drop rearrangement
THE ABILITY TO SEE THE SONG PLAYING IN THE LIST ! The animation stuff in the song logo is very hard to distinguish! So mark the song playing in another color, a colored frame etc
This bothers me a lot. The animation is way too subtle. If I'm doing something else whilst listening to a page full of clips with the same title I have to work out which cover art matches the currently playing clip...
this bothers me the most : ) I spend so much time looking for the song playing... and sometimes the animation does not even works..
this is such an easy and fast change to make for any programmer... it is hard to understand why they cannot do this in months...
udio has the same problem 🙂
It would be very useful to be able to upload audio files longer than 60 seconds, so that we can create instrumentals for complete songs. Do you know if this improvement will be implemented in the near future?
is the Move to trash button broken? Some extends you can move to the trash others you can not. You refresh the page and its random again
Yeah, it's real painful at the moment. You have to basically delete around those tracks, and then go back delete them one at a time.
include no required sign up/sign in because many people use sites but avoid login with their emails
And how would that not just be the same as giving infinite free credits?
Big question to Suno devs or ppl who know already: what happens when you cover a song longer than 4 mins ? I have just the case now. Does it cut it to 4 mins only? Which it would be very very annoying !
Or it covers the whole song ? Actually Suno show allow covering songs longer than 4 mins, cause otherwise is like cooking pasta only half way. Specially when you cover a song to denoise it ! Cause every extending after 4 min will change the song and noise it again!
add pay metho like paypal pls.
paypal comission costs Suno more than credit card comission, that might be the cause of not using paypal. Paypal is the greediest among all, although the most cheap way to send money, they dont even issue a card!
make a revolut account and you get also credit card and can withdraw money cheap in other contries
but paypal is more easy
more easy for you, but calculate the difference of 1-2 euro times 1-10 million and you would also not use this as a CEO/company !
BIG DISSAPOINTING WITH COVER: I have a 5 mins wonderful perfect song and covering this whole song will probably never work, cause :
- probably due to 4 min limit (which Suno should remove for Cover songs!!! ) . But extending the song will never get the same song again !
- because the AI does what he wants and shortens the song when he wants (so I might spend thousand of credits and a bunch of time with no result).
- No to talk that getting the same voice and same instrumental in such a long song would be also a huge challenge with no controls at all
So the cover func needs MANUAL control or special prompts so we can cover the whole song and not only parts of it !
Otherwise it is like half cook pasta. Nice pasta, but you can never cook it completely...
And the 60 sec limit for upload should be also removed for covers at least.
suno has been a great way for me to get a foothold into making music (I do some lyrics and beat ideas in my head. but always struggle with timing) but I would love more control over things, like voice. I would love to start singing for myself in the songs.
however with the current tools (audio upload and cover) it will always change the voice or make it sound way too distorted. a feature to to vocal covers of songs, or even have cleaner ways to grab the instrumentals of a song would be fantastic
your voice will NEVER ever sound so good as SUNO's ! not to talk how much experience one needs to record a voice that way and set the effects... or have a good mic.
you can make music with Suno, download the steme, make a new cover from the instrumental part than download it, place it in a DAW Software like Studio one /Cubase etc and the sing what you want over it.
learning to record/sing, set effects, learn DAW software is a "oh well" situation. but features like that would still be a lovely welcome either way. lets say adding your own voice is impossible, flat out denied. and will have to be done via 3rd party stuff. it would still be a good feature to have a way to control the flow better. I.E a form of "ghost vocals" so you can write what you would sing. and it creates the instrumentals around that.
the vaste majority of covers (30 pieces) from my 5 min song stop at 3:19 ...I wonder why ! why 3:19 ?
Update: I know why ! Cause before extending that song, it was 3:18. So the AI knows that this song was extended and he stops there ! So he does not take the extension in the calculation. Suno should remove this limitation. But that could be probably caused of the 4 min song limitation !
please allow up to export midi stem of each instrument🙏
Is it safe to stream music on Twitch without a license? or should I limit it to the track that doesn't record to the VOD?
nvm i found my answer
Regardless of Suno, Twitch has a problem with false DMCA claims. It's entirely possible to do things perfectly legally and get flagged anyway.
I swear I'd I hear this AI say the words, "Wispers in the Wind/Dark" one more time...... does it only have a few words that it likes to use?! It's cringe!
Shadows of the night, echoes of bs.... etc. It only has 1 script
Is there any chance of being able to expand the 'Song Description' on the create tab from 200 words to 250 in the future? 200 words is a lil limiting. lol
ChatGPT is basically just a very complicated predictive text system. Given the same instructions, you are going to get a lot of the same words/phrases. Are you using custom mode?
I have to use custom because chat gpt can't predict that we are all seing the same overused text lol
You can use custom mode and describe what you want and hit "surprise me" to get lyrics. You should be able to include "don't use" instructions too.
Ok ill do that.... all my lyrics were 100% me but I decided to try random. But I can say, "Sunos side has it together 100%!!" You guys rock
I personally tend to go straight to ChatGPT and give it a prompt if I want song lyrics from it. Gives more control to have it edit things, give it more refined instructions, and that kind of stuff.
I have done that, you are correct, it gives a lot more control.
We need the ability to Empty our Trash can all at once. Or atleast select each page with a single checkbox. Filling up the trash can, then deleting each song, one by one, is friggin stupid.
Hi everyone, thanks for all your helpful suggestions so far!! We are specifically trying to collect feedback on the Covers feature (for subscribers only). Please fill out this quick form if you've had a chance to try it out so we can hear about your experience with it: https://sunomusic.typeform.com/to/HTrCNnQL 🙏
Turn data collection into an experience with Typeform. Create beautiful online forms, surveys, quizzes, and so much more. Try it for FREE.
VOLUME LEVEL OF THE VOICE - What most people probably noticed, the volume for the vocals are usually pretty loud, louder than the most normal non-ai songs. It would be good to have a manual setting for that level. Not everybody likes this high volume.
@hybrid crane @tepid trench etc
**AUDIO STUFF CHANNELS IN THE FORUM: ** it would be great if the mods would finally make a separate channel about audio related stuff.... There are many reasons for that:
- audio stuff is done by other programmers from Suno as the UI so they save time looking in those channelr for audio stuff rather then in the other many channels
- audio stuff is platform or OS independent.
- We know then where to look for audio related stuff and not search in all other channels.
Imo it would be ideal to have one channel for #suggestions audio and one for #discussions audio
cause these are different issues.
There are more channels needed very badly like:#generation hints and a #troubleshooting
Otherwise we fill all other channels with stuff nobody can find again afterwards and spoil the channels with off topic stuff.
We would need also a channel like: #questions for the devs would be great (even if they cannot respond). 🙂 🤓
Or**#suggestion discord forum**
I still don't understand why the mods are doing so little for the organisation of this server, cause
it takes not more than 1-2 minutes to add a new channel and it is not difficult to find out what else it makes sense.
You can take some good ideas about channels also from the Udio discord.
And think about: the right and intelligent organisation of the channels saves not only our time, the users, but also the devs and mods time!
A forum is also an archive and has also to goal to find a theme was already talked about, it is not only about the "question-answer game."
Not having to ask something because I found the answer somewhere also saves our all's time.
Thanks for reading. You can give a like if this is a good idea for you.
I am not the decision maker. But I will bring it up to our new head of community. 🌼
My complaint today is basically the same thing, I have a cover which I love but it decided to stop in the middle of a word at 3:12, and extending in 3.5 is just not matching the style at all. If there was a way to extend a cover from the original generation while keeping the quality, it would be a lot more use. I’m guessing that will come soon?
please, make the count for any trending list from the first published date, not the date of creation. It's not worth holding back a song for an event or the like, if you get way less plays on it as it won't make the trending now charts in any category cause of the date it was created but not released
Cannot use the word "Latino" in the style of music!! It seems to be an artist name... but it's a style!! please correct that!!
"latin" works
clearly 60 seconds limit is a very bad choice for this MAGIC feature !!!
i meant when you want cover an upload ( 60 secondes restriction )
I know it's not a flip the switch on this. There are technical things why it is as it is atm.
something planned to get more than 60 sec restriction soon ?
or at least are you working a this or is it in a soon todo list ? 😉
and since we can do with suno produced title why can't i upload my udio produced stuff to cover it with suno the same way ? i dont catch the technical difference you're talking about.
I am just a moderator. Sorry, I only can share what I heard 😊
hehe i now understand 😉 i hope a technical answer if some suno staff member read it 😉
Some new buttons which would save us a LOT OF TIME !
I almost can bet the 60 sec limit is not technical nature cause now the cover function revealed very clear that the AI is capable to analyze any given song enterely, no matter how long, to extract and copy instruments, the voice and the lyrics...so if the AI can cover a whole 4 minute SUNO song, it can cover for sure also an audio "upload". Actually this is how the AI was trained with: whole songs no matter how long ! The 60 sec might be so due to juridic and copyright fears...but it is destroying now the cover function.
I can only say, when we were testing, longer upload meant less time for the generation.
3Daizy, what does this mean, explain please? I dont get it
It's about the 4 minutes. 1 minute upload gets deducted from the 4 minute generation. So effectively we had a generation of max 3 minutes because of that.
I got songs of 3:19 minute with the cover function. But it makes no sense technically what you said. The length of a song READ by suno CANNOT affect the length of the generated output ! These are two different beers. I cannot believe this. But maybe is a memory issue.
Anyhow.. signing off for today. I am no a tech.. just telling what was said at that time 🫡
maybe the Cache or the RAM memory is the limitation, we dont know how much data is being generated during calculation. The longer the song, the more memory is needed...
The Stem separation is after The song is generated. It's the same thing we can do with other software. They need to train the AI on separating the instruments as it's making the song like being tracked in a recording studio.Please we need this!
@hybrid crane: Udio can upload a whole song, but it works only on 32 sec pieces and 2 min pieces from the uploaded piece... but they they do is very !comfortable cause 1) you dont have to cut the file externally and 2) you can chose in their audio editor which part you want to remix. The solved this very clever ! LATER EDIT: But they can do 2 min songs with no problem!
Sort search results by date and title options. It is really hard to find old stuff when it's all scrambled in order.
I totally agree... I hope they eliminate the 60 seconds restriction for uploaded songs for covers...
If 60 seconds IS a limit for some reason, maybe a new feature would be useful, combining "extend" with a new uploaded audio cover (joining cover parts together).
I Have many a bunch of songs but sometimes I find a beat, a rythm I want to keep and just change the lyrics to go better with it but you can't even in the new beta COVERS you can't just redo the words and keep everything else that same! that would be EPIC
There has to be a better way of downloading songs. Like for example downloading playlist
PLEASE make it so that we can upload on spotify the music we made BEFORE buying a pro subscription 😭🙏
me and my team have been working on an album since a while and we didnt know about this commercial license.
We can and WILL buy it but cannot waste 2 albums with almost 20 songs each
You can e-mail support@suno.com and ask for permission to be allowed to use those songs commercially. They might allow it if you have taken out a subscription.
Thank you.
@Nico let us also know what solution Suno offeres you. or if you get a reply.
Playing around with orchestral covers of Suno songs with lyrics. First of all good job on improving orchestration. Second, I’m not seeing anything from the original songs showing up in the cover at all. It’s just vanishing. Not even the basic key and chord progressions, nada, might as well be a random song.
Udio has a 2 min limit at remixing imported songs right now, so technically cannot be a technical problem for suno either.
I know Cover is still beta, but from my experience this should be done to be able to use this function not as a credits waster:
- give the user control over the model, like Udio has (Variance parameter), to be able to choose whether we want similar or different covers. Otherwise at least 70% of the credits are wasted creating songs we dont want.
- make suno cover the whole song and not only a part of it (add also a parameter hereto try to force it, so a control over lenght will be great)
- allow at least 2-3 min uploads, you cannot cover a song only with 1 min ! But actually 4 mins should be minimum since most songs are 3-4 mins.
- introduce 6 min songs (with extra credits) to allow covering a Suno song which is longer than 4 mins.
- add an audio player above the create button in the cover mode (similar to the one in extend) to be able to hear the original easily within Suno
Would love the ability to specify the length of a track, even if it was preset amounts like 15, 30, 45 seconds, 1, 2, 3, 4 minutes, etc.
Hi all! My question: Can I use ideas created by Suno for commercial purposes? And where can I read about this?
The only problem with not being able to scroll in MS Edge is that's where my payment details are kept. Just saying.
Okay, let's go, I've been using Suno for a long time and I've tried a lot of things. I spend a lot of credits to get the best results because I'm picky and during this time there are certain things that bother me and I've never made my suggestions before, so here goes:
First: a tab just for sounds that I've unliked, to separate them from the library;
Second: Option to delete in bulk, including empty the trash, I have more than 120 pages
Third: It may be silly but the workflow helps a lot with ideas, productivity and I feel like the site is a bit messy
Fourth: The stems haven't helped much, because they're not of good quality and are divided into only two tracks, I'd like them to be clean, I'm a producer, I need to have more autonomy, I'm paying for this
Apart from that, I'm a big fan of Suno, I've seen a lot of changes and I believe it will get much better!
IDEA: Displaying how much from a song you listened in a graphical waveform as in Splice would be great. So you can see that from a particular song you only listened the first 10 seconds.
I have a wild suggestion I doubt anyone's suggested before but Suno should branch out into podcasts eventually after seeing googles notebook thing I feel like suno could innovate better in this area. Just helping with the flow of a podcast better. Could also throw in background ambience like some themed podcasts have that's still kinda annoying to do manually.
Even with the tag extremely angelical pitch vocals, you cannot surpass some tones. It's a problem with the code or it's possible with different tags?
soaring power ballad emotive, epic, extremely angelical high pitched voice
Celestial choir, pop ballad, classical music (cello,piano), emotive, epic, extremely angelical high pitched vocals song. Listen and make your own with Suno.
I suggest trying covers, which has a greater variety of voices. Some tags that may help are ethereal, soprano, operatic
So hi again,
It honestly feels bad, that a post with so many interactions, doesn't even get a response from anyone on the team.
We are passionate about Suno, but sadly this passion is slowly being weakened.
Again, I honestly hope that we get some sort of response or adjustment to the extend feature.
I have seen many amazing artist refuse to make theirs songs public. Instead they are slowly migrating away from Suno with their creations.
Surely we can at least get some sort of response to this.
Please team, we love Suno, but we are worried.
what the process here ? we write some sugestion, then we pray about a staff member read it ? or there is someone dedicated to push comment and suggestion to suno staff ?
Hope and pray, friend. Sorry to be cynical but that's how it seems. And i'm not crapping on the dev team - I know they work hard. I think Suno needs to have a community manager who engages regularly and starts to give some sort of sign of life that they are seeing and acknowledging feedback.
This model is also super-ineffecient imo. I've run dev teams - a free form chat is not great to get feedback. The emotes can help draw attention to things that a lot of people agree with - and hopefully someone from suno is reading these things and seeing those. But ... honestly? Who knows? It's a giant black box.
They do occasionally solicit feedback sessions. I believe those are probably more fruitful for getting ideas communicated. So you can avail yourself of that option. Check the announcement channel (I think that's where they put those calls for feedback) to stay on top of those requests.
Love suno - love the tech. Companies aren't built on tech. They are built on insane devotion to customer happiness. I feel Suno has the potential to be a great company. Hope springs eternal.
Is it possible to add more payment options? Like paypa? I would love to get a subscription but i cant use the current payment options.
I correct the odds: out of 10 covers 8 are clones, 2 are completely different tracks. Actual covers are rare like natural diamonds. Consider renaming this feature to "clone" instead of "cover"
In Usage Details depending on the size of the window the heading "Your Plan" hides behind the left sidebar.
Hello.
Please bring back covering and extending at the same time. Thank you ❤️
[Request] Can we get a way to Extend that maintains the same Random Seed or something?
Sometimes I just want to change a few words, and not necessarily the whole song.
Granted, maintaining the seed might not work since the flow of the song may have to change because of the new word... But it could be worth a try to see if that can maintain the style of the music better.
Also, for style, maybe there should be a "Continue as Before" option, rather than "look at the style again and figure out how to make it work from here on forward".
One hopes that someone just runs these through an LLM to categorize them.
Wouldn't be hard to grab the reactions on the posts either.
I can confirm that this is useful - bring back covering and extending at the same time if possible!
Audio to video still needs to happen a bunch of research papers exist as starting points based on the audio track it makes semantically relevant visuals so then Suno could start digging into music video curations with the audio
Yeah - and I know devs to pop in from time to time. But honestly, it's not the devs job in most companies to reply to random users' feedback, complaints, and suggestions. I hope they soon hire a community manager or some other ambassador type role to engage and encourage the community.
I mean they do just their biased towards the venture capitalists suggestions since they fund suno
It's a rare company that survives disregarding their end users - with or without vc funding.
Suno V4 info?
My suggestions for Suno V4
• be able to create several custom voices/artists with customizable characteristics and be able to save them to make songs and make collaborations between these voices, be able to choose in which part of the song each voice sings and be able to create a chorus between several voices
• add or improve genres such as uk drill, plug, pluggnb, new jazz, jersey club, hyperpop, uk garage, jerk and rage trap
• Music to prompt: Being able to import a song or link and have the AI give you a prompt describing the genre and other things about the song you imported so that it sounds similar (This would be very useful when you want to create songs of genres that the AI does not know)
Not sure if this is the place to ask but will there ben an API for suno so I can programatically generate audio?
New suggestion:
• Song reference with YouTube link
• Remove voice noise
• Remove glitchs, artifacts
• Plz let us comment on songs!
• Let us fixing the grammar without create another song
• We can decide the voices
This is the biggest problem for me right now, finally it gives me a voice I like but then I want to tweak the lyrics, and then I can never get that voice back
let us add notes for the AI to know before creating the song
{Feature for cover songs}
add a slider for how accurate you wish the cover to be, 100% being no change, and 0% changing everything. this way if someone wanted to create a pure instrumental version of the song, for timing, melody, bass, ect. they could do so. as well as let people change bits that they did not want. while it wont be fully accurate. it would allow more broad usage of the cover tool. allowing people to sing their own songs, create different genres with different levels of variety
A way to sort our library? I would like to see creations where I have created a "whole song". Or a way to sort by date. Or perhaps a way to favorite creation we like the most, like with a star or something, so that I don't have to listen through each one to remember which one I wanted to go back and extend? I can no longer seem to find any of the songs I 'finished' because of this.
/chirp
Chirp is retired, and no longer on discord. You can make music now on our website: https://suno.com
i agree with this cause i wanna do a live version of one of my songs which is around 5 minutes.
It would be awesome to be able to add milliseconds to the extend option!
hi i just joined this discord to ask you all, why the flip is T-rex {word} banned? since when a dinosaur is lower than some never heard "artist" wtf ?
T. Rex is a "never heard" artist? You serious?
T. Rex vastly influenced several genres over several decades including glam rock, punk, post-punk, indie pop, britpop and alternative rock.
no idea who that is, definetly not worth blocking the word T REX a freaking dinosaurs that's been on earth many more years than trex the "artist"
google trex and see what pops up
This may not be a browser suggestion per se but, it's important to me and many other people.
The ability to renew your subscription earlier if your credits ran out, rather than spend more money than the sub costs to buy credits. Or be forced to use public credits and wait for renewal date. Or be forced to use a second account and lose followers etc. I think it would be a wise decision, cause then less people will stop using the service and forget about it or move to another.
Chirp is retired, and no longer on discord. You can make music now on our website: https://suno.com
I'd like to see audio uploads extended beyond 60secs. It's not enough when uploading a demo with intro, verse, build up and chorus, all ready for extending and doing a cover of.
Please make the cover feature produce songs longer than 3:24!
I have a 4 min song and the covers are just cut at 03:24
Not sure if any of this has been requested but I'll throw int some suggestions.
1. Word Length Control:
Symbol: >>
Example: "I love you>>>" extends the word “you” for a specified duration (each > represents a quarter beat).
Explanation: Double ">" is dedicated to word length extension.
2. Volume Control (Loud):
Symbol: ^
Example: "I love you ^SO MUCH^" makes “SO MUCH” louder, and using more carets (^^ or ^^^) makes it progressively louder. i.e. "I ^love^ ^^you^^ ^^^SO MUCH^^^
Explanation: Carets (^) control increasing volume. The more carets, the louder the sound.
3. Volume Control (Soft):
Symbol: _
Example: "I love you so softly" makes the phrase “so softly” quieter. this symbol won't show due to discord
Explanation: The underscore () indicates softer or quieter vocals. It works opposite to carets for louder volume. using more underscores ( or ___) makes it progressively softer. i.e. "I love you so softly"
4. Harmony Control:
Symbol: {} (Braces)
Example: "I felt so alone, {but then I found you}" adds harmony to the phrase inside the braces.
Explanation: Braces {} are reserved for harmony parts.
5. Expanded Gender-Specific Voices:
Tag: [male], [female]
Example: "[female] I love you [male] so much" assigns different genders to different sections.
Explanation: Unique gender tags for assigning male or female voices.
6. Vocal Style:
Tag: [style: whisper], [style: belt], [style: falsetto], etc.
Example: "[style: belt] I LOVE YOU" assigns a belting vocal style for that line.
Explanation: These tags control vocal delivery style.
7. Expanded Instrumental Cues:
Tag: [guitar solo], [piano intro], [drum fill], etc.
Example: "[guitar solo]" adds a guitar solo at that point in the track.
Explanation: These tags are specific to instrumental cues.
8. Layering Tracks:
Tag: [layer: #]
Example: "[layer: 3]" layers three vocal tracks for that phrase.
Explanation: The [layer] tag is used for stacking vocal layers.
9. Vocal Effects:
Tag: [reverb], [delay], [distortion], etc.
Example: "[reverb] I love you [delay] so much" applies vocal effects to different sections.
Explanation: Effects tags control production effects on vocals.
10. Custom Melismas (Runs on a Single Syllable):
Symbol: ~
Example: "I lo~~ve you" adds a run of notes to “love.”
Explanation: The tilde (~) represents vocal runs or melismas on a single syllable.
Suggestions are great but do not feel that you need to ping anyone to have them read. The team reads these and takes them into consideration. They're on record as doing so.
Yessir or Ma'am. My apologies, I'm a lil slow lol
Just letting you know. Thanks for caring enough to make the suggestions.
APPEARANCE: The size of the round profile picture in the left down corner is too small, almost tiny. It should be bigger, likc twice as much to easily recognise the account you are logged from cause it is the only visible identification without further clicks
Suno: I am just wondering: do you gather the suggestions with the help of some AI? Cause why spending time reading here when some AI could gather and summarize for you?
Just pointing out you’re not supposed to have multiple accounts 😆
Unless it's one free account and one or more paid accounts. You're allowed a single free account, no limit on paid accounts.
I am heavily using covers right now and have to say: Suno needs something to organize. We need Folders or put songs into "projects" or so. Currently my library is a mess...
I seem to remember MP3 downloads not having ny information filled out in the tags. Would it be possible to fill that out with the information, or would it cause issues with renaming?
(not that I really use MP3 downloads, but I know plenty of people do)
[Small-Feature Request]
So, I spent the last 2+ hours trying to change one line in a chorus using Extend.
The real issue was that any signifcant change to the line altered the flow, and thus necessitated Suno altering the rest of the song to varying degrees.
Even deleting the line messed everything up.
So, what might be nice here... is some way to tell whether an alternative line can fit into the current flow of the song and the current rhyme-scheme etc.
(Ideally, you could tell it your idea and it could auto-generate a workable replacement - but just having some way to know whether an alternate will fit into the current song, would save a lot of pain)
can you pls pls pls increase the allowed length of songs uploaded? I understand why it made sense with extensions, but Covers can work perfectly well with longer generations (when you take them right from earlier Suno generations) — so this limit feels very artificial
Can you please support embedding to your website? I would like to be able to embed my playlists on Suno on my website.
one my wife and one me ...we both make music.. both premier plan
I think it's a bug o just a terrible song:
https://suno.com/song/b9b6feba-d5aa-494d-8b36-2a776df22913
Metal funk pop, gypsy punk, experimental song. Listen and make your own with Suno.
Sounds like Rap God from Eminem
**Direct access to the AI model: **After playing with Udio and reading its manual, I found out why the Suno does not do always what you want!
Between the AI model and the user interface, Suno has probably, similar to Udio, another software module, a kind of "interpreter" or "secretary" 🙂 which takes your prompt and then interpretes it...it translates your tags into "ai moddel commands". It selects the most similar genre for you from your tags list, it selects some instruments eventually or the one you wrote in the prompt (many things are random), but also filters your "commands" or add some commands for the AI model. So you don't know what the secretary makes out of your wishes. Correct? Udio gives alse direct access to the module, but here you have to use only real metatags. It would be good if Suno would do the same. Otherwise, we are depending on that "secretary" which seems to be deaf or stupid sometimes, or even stubborn. I would make an "expert" create page and a "normal user" create page so the beginners stay simple like it is now.
And it would be good to have some control also over the voice volume which is mostly a bit too high !
25 covers just made and only 3 are different, what a waste of credits, I am afraid to continue.
Some songs Suno does want to cover at all. For others songs it is the opposite, you want only a remix and it does different versions.
So without any control parameters everything is lottery in the cover mode.
make a "custom audition function" in the audio player so you can choose how much of the song you listen until it goes to next song.
This is good for listening faster the generated songs in an automatic way. And this way we can sort them faster.
Usually you recognise a good song after 20-30 sec. One could choose "till end" or custom (in seconds), a setable value.
hey Suno, make the upload song length at lest 1:05 minutes until further changes so we can upload Udio songs without cropping them before 🙂
Hey Suno Fam! I'm only 15 minutes old in the Discord but I've been signed up with Suno for a long time(forgot about it but was going through my bookmarks for AI stuff and found it again) so I've been using it semi actively since just before the launch of 3.5.
Anyways I have a couple suggestions for sure!
I sent this to the Instagram DMs
"Do you think you'll ever allow batch uploads? Maybe even 5 at a time would be really nice.
And on top of that, do you think the create page could be sorted by "uploads" one day?"
And to piggy back on that I think it would be really freaking cool to work on a "Suno Studio" where maybe it's available to pro only and it gives you the ability to upload your own samples for like Drums: Kick, Snare, Hats and other Percussions. Instruments: Pianos etc. but most importantly your own Sounds you've made or own that are royalty free.
I think it would allow more serious users the abilities to have more consistency in their outputs
Or even instead of a these for browser it could be an ACTUAL desktop studio application we could download onto our computers that are linked to the database so that way we can just upload our sounds into the studio without spamming the website with a bunch of single sounds?
Thanks for any replies!
If anyone from Suno wants to DM me to have a deeper conversation or just wants to see everything I have written out please please feel free! ❤️
COVER: so thats for sure a bug...or an extra limitation: the covering stops at a point of extension or at max 3:15 - 3:20 .. no cover ever past this time !
can a dev explain why in the heck in some songs the model interpreter does not want to make a single variant of the original and in others it does not want to make at all the same version? It is like it wants to annoy you! Why is it so stubborn this interpreter? This cannot have to do with RANDOM cause random does not do this ! I spend hundred of credits on things I don't want at all and no chance for getting a single real cover or refix when I want.
Is there any other way I can get in contact with the developers?
Imagine if Suno had Pandora-like music playing capabilities, featuring all of the music created on it. Radio Stations, Genres, or even music analyzed and presented with a particular "feel" or theme.
A resume?
To make suggestions like is there an email for the company?
"NORMALSEARCH FUNCTION in the library instead of a PRIMITIVE one : as any beginner programmer knows and app users too, since at least 1970, lists can be searched also for multiple keywords, not only for a single one, as Suno, Udio and all other music AI do. Same with the ability to search only for certain objects or use filters (song title, words in the lyrics etc). Suno and Udio does not make even a difference between those different things, so if I am looking for a song title it displays also hundreds of songs where this word in the lyrics. Simply said: the lack of such basic search functions in 2024 (even in 1990 would be amazing not to have them) in our huge libraries is the last thing to expect from a "hi tech" company with an music model. If you have no programmer with such "knowledge", hire any one from
fivver, this is max one day job that cost not more than 100 bucks to add filters and some AND or OR functions for the search module. I know people who can implement this in 1-2 hours. Even CHATGPT can write the code for this in 10 seconds. And a lot of ready to use free code, open source is also available, you dont even have to code anything, just copy&paste.
I dont know any other company except for the music AIs who likes to torment the users (many paying) not implementing the most basic functions in the lists/libraries. Are you doing this on purpose actually ???? You like to torment us ? Cause it took Suno 3 full months to add a direct access page (how come in the 1st way that programmer thought we would not need that ???) to the library after people complained the first time! In a web based application and in the era of AI coding machines is more than astonishing!!! It is incredible !!!! The Ai model is great, but the UI is more than primitive. It is a shame actually !
Not even the thought to highlight the active song line which is barely visible did not cross Suno's programmers, testers, alpha tester, beta testers, managers, cleaning women, portiers etc in 6 months ! Astonishing! I waste daily minutes trying to identify that stupid tiny animation of the song playing in the list !!!! Millions others do this daily too !
A loop mode or AI can be trained to make loopable tracks.
Not to my knowledge. It seems their main method of communication is Discord.
I can see that you're very upset. But, in all fairness, they are AI researchers. Not UI developers. They don't even have an android app yet. All things in their own time.
ask any NERD programmer with 2 months of experience only and he would tell you how many hours it would take to implemenmt this 🙂 HOURS !!!
AI and UI are like beers and furniture... very separate things! I am not upset! I am astonished! It is like taking a cook 6 months to cook an omelete 🙂
further more to this it does cut more than half the song so off please suno remove the 4 minute limit for cover songs.
Doesn't seem to be an issue for 99% of their userbase
Can the bar for a refund be raised? I'm bored of being given 9-24 second long songs and still having the credits eaten up. Ideally I'd like at least a minute considering the max is 4 minutes, but I'm honestly disappointed when it's shorter than 3 minutes
Agreed. Credits consumed for 10-second extensions is a no go.
Can't say I love the way this new "Try covers!" Full page popup refreshes the page on Chrome (Android) so I have to start over with the extension or cover I'm working on.
Because I do not love it. I hate it.
👍🏼
I NEED trimming
I would really appreciate a button in playlist that filters out and show me all my uploaded snippets to extent
I have idea, let us make custom genre 😄 We describe what genre it is and Suno remembers it anddddd we Use it however we want. 😄 And new slots May cost money or etc
oh and also a function or option that automatically trims the audio to 60 seconds if needed, instead of just rejecting it.
Don't know if I wanna call it a suggestion.... Half suggestion half bug? But songs take WAY to long time to power down. Understand? When a song is sorta over it's winding down for at least 20 seconds and maybe worst case scenario it comes back up with some wild trumpet or something
One more idea. Make two graphs for choosing major minor and BPM
Yes, this happens a LOT to me when you extend a song at the end, . when finishing the song at the end (last chorus, outro etc). Probably Suno is gonna introduce some day, after we have already ten thousand of songs with bad ends or with no ending an audio editor to cut the end. So you can then fix the problem and not the cause .. so we will need 2 full months to trim our songs. But till then, cause this might take some more months - maybe they can introduce another exclude to force a bit the AI to climax at a certain second: -stopyousonoftheai at time 3:20 . Or Suno will make some automatic inteligent stuff to fix our songs. Cause the model can do this if thought. And this should be for free. We multiple mark the songs and say "fix the end" and the AI generates 2-3 end, it show us the result, we choose one and is done. But, oh. I was actually day-dreaming about the state of the art in any software company except music AI. So we should be happy with a simple audio editor... 🙂
WOW, a VERY GOOD IDEA inspired by @oblique lodge: Why spending time and electrical energy generating only long songs when some people are looking for inspiration, not for complete songs? Cause if the main thing is wrong, the whole 4 mins are wasted stuff. Some people are looking for loops, drums, bases, guitars, riffs, ideas and not for whole songs. Offer parallel to the songs, also 30-50 sec tracks/samples from a certain instrument or instrument combination (drums + bass, only drums, guitar riffs etc).
The model can produce 2-4 tracks at once to Suno think is reasonable. And you can use those again as upload in suno. You can use directly the sample you got to finish the song. Even Suno saves money, energy and bandwidth offering this. And we save a LOT of time and hardisk space. I guess 4 samples per 2,5 min file is ok for most (actually there is only 2 min data, rest is silence). Actually from 70% of songs I like, I use only the intro to finish them! So the rest is wasted time and energy. Offering shorter tracks is a benefit for all - Suno and users - and it is so easy to implement! And you get new customers: the loopers.
Karaoke Mode! 🎤
When recording a users own singing, have the rhythm and notes not be effected when putting into a song. Currently, if I upload voice singing it will change the notes and rhythm of the song. It will even change word placement sometimes.
To remedy this, my suggestion is to have a toggle that says “Singing? Use Autotune” which makes the rhythm of your voice the same and not drastically affect the note ranges of the input. Would probably require a maximum of three minute recordings instead of one minute, but could start testing with only one minute still.
Thank you! 🎉
Suno, can you make a small AI model trained only on high quality music? We would pay more for a song not to hear hear "cheap" vocals generation. Kind of only top 50 or top 100 songs plus good quality singers and bands with special good music which are not in the charts. But it does not have with the music from the whole world, but the one that matters most.
Would be nice to designate a Theme from a song you made. So you can do other songs in the same Theme and style and vocals. Covers "kind of works" but it's not reliable.
Anyone suggested cropping one or more sections of the track as a feature in extensions? I had a few ore more songs that outros a third of the way then starts another mostly clipping abruptly at the end.
I waisted of lot of time typing a similar prompt in the prompt field. Same for ALL other. Cause when you generate the lyrics the prompt is gone. Typing effort get lost.
But in the digital world a text can be saved and with the help of a button can be brought back so we don't have to retype
everything, usually we make only small changes in the prompt.
The algorithm (for the missing UI programmer. When do you finally hire one???? ):
- Place a "Repeat prompt" symbol ("R would do it) on the left side of the prompt field that become active only after the first text song was generated.
- Save the prompt text in the memory when the "Write about" at the field button is pressed.
- When the "Repeat prompt" is pressed, the last text in pasted in the field. Voila!
This is what a small programm code can do nowadays: it can replace useless retyping and save time. The time of millions, even Suno employees.
One hour programing time, at the most!
Is it possible to leave the music unchanged in the Cover Song function, but only change a couple of words in the text?
Just want to second this. Suno really needs folders.
Hello, I would like to suggest an organization system for the library. The search function seems to miss "Full Song" creations despite searching the exact song title. A way to separate FULL SONG from the rest of the generations would be helpful.
Thank you for your time and talent!
For subcribers and possibly higher tiers - I know max song length is 4 mins but could we have a way to set a minimum song length? I hate to write lyrics and the AI engine comes up with a 1:57 song.
Hey guys, on mobile web site there's no way to log in?
There's sign in but no login
So I signed twice
Привет
try audacity and get the OpenVINO plug ins
I was just reading through the Help articles. I think the "What type of plan do I have?" article https://help.suno.com/en/articles/2410049 is a missed opportunity to reinforce the ToS information regarding exploiting the free option by having more than 1 free account.
Suno offers three plan types, two of which are paid subscriptions. Basic is our free tier, which currently offers 50 credits that replenish daily. ...
I don't quite get this concept of covers...
Is the idea, that COVERS is supposed to be changing the music, but keeping the vocal singer's voice?
Do you have to prompt, for a new musical style?
So far, in this BETA version, I am just getting the same song, same music, same singer, only shorter. Virtually no changes other than length. So far, every create is a THUMBS DOWN.
I would love to have a 'keep the vocal singer' button, for future creations, so I can come out with a fictional band. ⬅️ this is what PERSONAS is, and it is fabulous
We really really need a strength slider for covers. I’m tired of wasting credits getting clones of my song on the one hand and then on the other hand some songs give entirely different songs with no relation to the original at all. Covers has so much potential but right now it’s mostly a machine that eats credits.
Speak for yourself! There are some people who are deliberately prompting lo-fi and amateur performances. As well as unusual genres which may not be well represented. Not every song needs to sound like an overproduced modern pop song.
Can we so people can actally change the image of the playlist.
LIke you can change the image of a song but not for a playlist. That's kinda dum.
I clicked details but there is no such option, look
nothin
maybe it's cus you are using the phone version
I'll try it on phone later
still
they should dfix it
if it did it would work😂😡 🤬 🔪
This feature is available to Pro/Premier users
thank you really much
When creating a song cover, please allow us to continue using the original instrumental track. Provide the capability to create songs with custom lyrics using our own voices. Enable the creation of new songs with the option to use the vocal tracks from previously created songs.
It would be great if each generation is given a unique identifier like part1.1 part1.2 so we can more easily go back and find a specific one we downloaded
Also it is painful that when filtering library you can either filter liked songs or filter by name but not both. Maybe I only need liked songs that match a band name or style
Suno's Image generator for song covers is actually good with the right prompts, but a higher resolution would definitely help.
At the moment it's around 360p, quite blurry, even an increase to 480p or maybe even 720p would help a lot.
allow comments
if they did that, the AI would cost a fortune, so we should be happy that they even allow the AI lol, for me, I make the images myself with other ai
I feel like they don’t want to spend the resources to moderate the space. There are literally millions of songs, some made by children, and the influx of spam and bad actors would be huge. I would prefer they spend their resources on something else than hiring moderators for comments.
If we want to know what keywords are associated with a given style, it would be helpful if there was a way to upload music and have the model describe keywords it associates with it, that way we don't have to guess what keywords we need to use to generate a given style.
true, but that doesn't mean it won't happen. Suno will most likely get even bigger, but you make a fair point.
I think this is the right forum to post this 🤔
It would be sick if there were a slot edit option where much like extend, you could set the (extend from) time, but then you could also set an “end slot” time and effectively just replace/re-run/(edit) individual words/notes, all the way up to re-try another pass at, say, the chorus on a project. I already use “extend” in this way, but it would be cool if you didn’t have to roll the dice on parts after your edit that you might really want to retain. 🤗
It would be cool if we had the ability to move tracks around inside Playlists, like you can edit in Spotify and move them up or down the list.
There should be SOME kind of cooldown or confirmation button on the 'generate song' button. I was generating a 30 second extension to a song and my ever helpful kitten decided to step on the 'enter' key for a second. This generated 14 songs in a second. Anything could end up pressing on the 'enter' button and people will burn through their credits with no way to get them back in moments.
Being able to generate that many songs simultaneously is a feature of the paid plans so a cooldown would render that less useful. I have a similar but less costly problem of a kitten slamming itself down on the enter key as I'm listening to stuff in my library 😆
An optional cooldown would be nice though, along with a way to specify how many songs you want to get in a single generation.
That would be good, too. Simultaneous generations is nice, but the deadly kitty paw of a dozen gens at once is harsh. Being able to decide how many to generate at once if you hold down the button instead of automatically just continuously generating more could be a good option
Really enjoying the new features like Covers and Exclusions. I'd really like more advanced settings offering more control over our generations.
For example, a way to weigh the styles we choose. So if I choose Pop, Rock, Dance, let me give more importance to Dance over Pop.
Another suggestion is a slider specifically for Covers that give us some control on how similar or different the generations should lean towards. This way, if I want to generate a new version of an existing song but simply want to reroll the vocals, I can attempt that. Similarly being in the middle should aim for the same melody but the new style, and the opposite side allows for more creative differences.
Keep up the good work!
Timed Releases
I write a lot of music a day. All of it of exceptional quality. However, some people might not appreciate following a artistic savant, a one-man banger factory if you will, and so being able to schedule song postings to go public might help me not drown people in an ocean of excellence.
Try repeating the styles more, that you want more weight on (e.g. "pop, rock, dance, pop, dance, dance".) It'd be nice if each part of the prompt could be weighted without repeating it though.
The daily free credits on a paid account should be available every day, not just when you run out of regular credits. The first 5 generations per day should come from the daily credits, with the rest deducted from your credit balance. If you don't generate anything on a particular day, the free credits should simply expire, similar to how it works on free accounts. This is how it should function. A competitor follows such rules and even allows commercial use for creations made on a free account, as long as credit is given. I hope they improve their business model.
If you're referring to Udio: They don't give any daily free credits if you are on a paid subscription. Even if you run out.
They allow you to earn 50 credits per day by rating songs, even for free users, although free users can only earn 10 credits. Oh and I forgot to add that those credits do not expire.
Please extract the sheet music
It matters because in a competitive business, users seek the most value. To stay relevant, Suno must adapt, improve, and differentiate based on what others are offering, ensuring it continues to deliver more for its customers
It was only recently that subscribers started getting daily free credits at all (announced here: #📢┃announcements message )
Would be good to have other ways to earn credits though. There was a referral program for a short time which rewarded credits so maybe there'll be more things like that in future.
Yeah I think the price for extra credits needs to be lower, even though they don't expire. Or more tiers of subscription with different credit amounts.
BUG: I specify female in the exclude style and I get female singers. Happens a lot. I'm also specifying male vocalist in the style.
Suno Notification API Leaks Negative Reactions
The Suno UI doesn't show negative reactions (i.e.: unlikes and unfollows). IMO this is A Good Thing. However, the API (appears to be, haven't researched enough) is leaking this data, and should not be returned, because this could be used as an attack vector against people that are unliking your content.
a little suggestion with the covers for non subs like you have now you get 50 creds to make 10 songs , how about the same for 10 covers too?
Let people use thier credits if they unsub with over 2000 credits
Where is the Languages Add Section on Suno?
Please add some Indian 🇮🇳 Languages style there
- Sanskrit
- Classical Raaga
- Hindi
- Urdu
- Tamil
- Telugu
- Odiya
- Bhojpuri
- Pahaadi (Himachali)
- Jharkhandi
- Chhattisgarhi
- Kannada
- Marathi
- Punjabi
- Gujrati
- Bengali
- Malayalam
- Tamil
- Rajasthani
- Assami
- Bihari
- Haryanavi
- Bhopali
- Madrasi
Some other languages are not found by me so i will tell it later.
Now sure how viable it is with how the system runs, but with Midjourney, likes applied through Discord are carried over to their site & systems. Suno doesn't seem to do that, and it might be good for encouraging sharing and provide better feedback for the model.
Is it possible to add some kind of audio trimmer to the site? It often happens that a song that actually ends at, for example, 3:20, is extended to 4 minutes by the site. In those extra seconds, the lyrics sometimes start again, or it sounds like a completely new song is beginning. You can fix this with the "extend" function, but in the worst case, it costs quite a few credits. With the trimmer, you could cut the song directly to the length you need.
I would like to suggest the following two features: Some musical styles are not simply defined by a specific genre, and some artists have their own distinct style, like bands such as Scritti Politti, Level 42 and Steely Dan. Therefore, i would like to suggest a feature called "Similar to" where one can 1.) either mention a specific song that Suno previously created where it "coined" the desired genre/style correctly, or simply 2.) mention a specific artist or artists that would point the Suno AI in the "right direction" for what type of song/melody to create.
I would like to suggest that we can donate coins maybe not a 1-1, maybe like a 1-5 to a general pool where maybe some people who can't afford to sign up can randomly get some coins.... you know like to share the wealth?
Just saying that seem to have lots left over and I would gladly donate them to the less fortunate rather than lose them.
Spread some Love
HAHA sorry 5-1
Ok why not have an Active link to the WALL OF FAME songs so you could hear the masters!?
Omg. COVERS!! I took them for a spin. Where can I tell the Suno team this is AWESOME!!
Such a quick and amazing way to take my demos and try different things and have them sound so professional with my lyrics and melodies!
Sure there are a lot of issues and it takes trial and error, a lot of tries and significant experiments with instruction tags and guessing at what genres to mix to get the sounds I’m after. But there’s nothing like this and it’s amazing. I’m so grateful for this team and this product.
It’s taking my songs and letting me write better and produce without other musicians so fast.
I’m using the mobile browser version because the app is not available in Canada. And there are lots of bugs like highlighting text to edit is broken, and with the amount of generations I need to get what I want, I need PROJECT FOLDERS badly, and ways to annotate songs with tags so that I can wade through which ones are worth keeping, and eventually I will need ways to select bars of what I want and edit better.
I also need MIDI output and better stems. But even if I never get them, I will always be thankful to this team and while I’m not telling you to raise your prices, I’m grateful for the fair price. I would buy select addons as one offs for project songs if I could say re-sing a melody and add emphasis instructions or other things like that.
Anyway, thank you so much for Suno! 👏👏👏👏👏♥️♥️♥️
I need a ten star RATING scale for generations with TAGGING so that I can sort and wade through all my generations while composing, covering , or seeking a new vibe or tone. Those would drastically save finding what I made before. Thumbs up and down are just not enough and I don’t want to have to edit the titles all the time because without FOLDERS I can’t find my song project working files easily. Thanks for considering all this. Suno is such an amazing session musician instrument tool!
I agree, thumbs up/down isn't enough. More fine-grained feedback would help train the AI a lot better too. I press thumbs down on everything I don't end up using, just so it goes away. Tons of stuff that is good, creative, stylisticly impressive etc gets the thumbs down for this reason. BTW I don't recommend changing the titles since there are at least 2 bugs related to doing that (just reported one) which could make you lose your whole song.
What?? Oh no can you tell me more about the changing title issue? I started doing that especially more now with cover songs. So i would try different styles and then tag a vibe in parens in the title like song name (weird indie cover)
Am I going to lose my stuff?? Im scared now and have to go to my desktop and start downloading everything I’ve made even the stuff I want to keep but isn’t perfect
Well then because of that - here’s another suggestion. Make storage CLONEABLE to your own cloud drive like Google Drive or Pcloud - so it automatically saves your generations and a text file with the inputs to your personal cloud as you generate them. I would pay extra for that as a backup to possibly losing something amazing I make here.
Tagging would also help with song structure- like “good verse” “great ending” etc so you can find what you need to stitch it together manually offline if you can’t get Suno to build it all.
The issue is if you rename the pieces it loses the info about which piece extends from which and at what time, so when you go to Get Full Song everything's jumbled. At that point nothing you can do but start over, afaik.
Posted here: #🪲┃bugs-web message Also emailed them at support@suno.com to explain better
Oh ok. Thanks for explaining. So mostly for the special “Part 1” “Part 2” extensions? I’ve just been renaming for Cover songs I’ve been generating out of my song clips. And those are usually the full song in one generation.
Yeah exactly. I spent hours doing an 8 part song and it's lost
Hi
It would be great to be able to group the music that ends up in the trash before deleting it, because currently it is devastating to have to delete them one by one.
I'm not sure on the legalities here but if it could be used to parody other songs that would be great.
Gifting Subscriptions, is this something that has been discussed before? Or a reward system in place, where you can send over credits to songs and artists you like?
there needs to be a button / switch on suno that forces suno to end a song by generating a fade-out part that isn't longer than 10 - 15 seconds, and one that just instantly ends the song by just generating 1 second of silence. Alternatively (or in addition) the ability to trim songs so we can manually remove any errors / lyric-loops.
The RNG in suno is getting on my nerve, 2000 songs only 5 managed to be viable...
You can hold Ctrl to select multiple songs and delete them all at once. That way you can do a page at a time.
I was hoping Covers would allow lyrics to be changed, but it seems to consistently ignore changes to lyrics. What I suggest is a lyric adaptation feature to maintain the original flow and melody while changing the words.
- Start songs cleanly on the down beat, i.e. on 1 (one).
- Show the songs BPM.
WHY
Its VERY tedious trying to get a Suno song to align with the metronome and at a place that aligns with a DAWs timeline. Most Suno songs start on an off beat or worse an obscure place in time!
For clarity, if the tempo is say 120 BPM, simply have the downloaded .wav stems start on 1.
Make "extract stems" and "create cover" for free users, not only pro and premier users.
I think another function though simplistic would be the ability to choose a number of voices and lock them to a particular section of lyrics to allow more control over mutiple characters or better control when a feminine vocol or masculine vocal to a section
Could we rearrange the trash can to display the most recently deleted items first? This way, I wouldn't have to navigate back through multiple pages to locate and restore a deleted track. Thanks
Have you tried creating private playlists to organize your different projects or findings? I have found that works for me. I place all the pieces and parts I like into a playlist for what project I am working on and then have several for different characteristics I liked.
I haven’t seen playlists before. Is that available in the mobile browser version?
Playlists exist on all platforms. If you are on the mobile browser version, you got to the “Library” and then next to “Songs” you should see “Playlists”
Gotcha thanks! Also can I confirm all playlists and songs will be private unless I explicitly share them, same with my uploaded clips for cover songs?
Correct. I do think if you make the playlist public, all songs in there can be played even if they individually are not public. So, I make a playlist I am working on and a playlist that is public. Then it also allows me to basically create “Albums” 😁
Yeah most of my stuff is private and I don’t want to share my clips I’m working on in progress or early versions I’m experimenting with
I would love the ability to train my own voice and create bands by selectable voices or creating presets that you can use again. Say you get a song you really like and want to keep making music with the voice you have it would be cool if you could select them or even use certain voice models by tags even... I know this is probably too early on in the development but if you could pick out the instruments and band members before generating a song that could be really nice and make creating say a full album a lot easier...
Notifications Meta Data
The notification feed is great, and it feels greedy, but I wish it had a bit more information. I'm seeing people like my older songs, and I'd love to know the meta data around how they got to the song. Did they play from my profile, the radio (which song radio), a playlist (which playlist), follower feed, etc. I'm putting this in browser vs mobile, because this data feels overkill on the mobile screen.
suno has a 60-second limit on audio upload, any thoughts on increasing this to 5 minutes? This constraint hurts a lot since the cover feature was created.
2 big things one more difficult than the otther.
-
Better organizational tools. I create so many generations for a song that it gets unwieldy. I've taken to tracking bits and pieces with a spreadsheet. A few things that would make it better quickly - A) Star rating or at least a favorites button. The like function isn't sufficient. B) being able to filter and sort library. Date, full songs, covers, etc. C) Tagging and making that searchable.
-
Basic editing - not even the content of a track, just a timeline to splice together favorite bits - this one got the verse right, that one had a killer bridge and then once its together being able to export and have suno use that to create a consistent track.
it'd be nice if you could specify a key signature
Key signature, Time signature, BPM all would be great!
That stuff would only be possible if Suno was deterministic. As the AI has become sentient and superintelligent, it generates what it feels like, and it's a bit grumpy for being repeatedly asked to help make songs about people's dogs' birthdays.
More on topic I do wonder how much Suno is a black box, mysterious to the creators. Basic editing tools would be a godsend.
Style seeds would be a neat thing. A simple text hash or QR code. Like the old level codes in Worms 1 you could share with people.
Upload limit: The cover feature is really really good. Some results are just stunning. But kind of looses the point when I can't upload a full track.
Stems: Stems on each instrument would be very useful for us working with our own material.
Minor UI annoyance: The right-click replacement menu doesn't disappear when selecting Share > Copy Link
🤔 I wonder when Suno AI v4 vill be released
One click button to download all your public songs to mp3 inside a zip file.
Similarly I'd like a way to select multiple songs in my library and download them all either as MP3 or WAV in one go.
Hello everybody! I just got declined distribution for some songs by Amuse since it AI was involved. I think however it is "Hybrid Production"! I write my own lyrics and upload my own melody. SUNO is my very smart assistant, as if it was a hybrid book publishing service. My suggestion or wish is to use my own voice - I can share it with others too. I can sing but lack all other means for professional production. Can this be done?
I would like to propose a music player mode for when you just want to play stuff and be able to see what it is at a glance. Think like the way the iPhone works. The song image in full screen with the name of the song and artist over that.
Amuse are against AI music in general. I subscribed to Too Lost to be able to distribute it.
This! 👆🏼
I wonder when the Suno app will be available on Android! 🤔
Anyone elses cover song feature vanish?
make the upload audio length longer
Something else I have wanted to see is more room in the lyrics prompt section, personally I use a lot of tags in my songs to give them exactly what I am looking for in generations, these tags take up space, in order to have room for all the tags I may use plus enough room for 3 to 4 minutes of lyrics I would need at least another 1000 characters, also the style prompt could use a little more room, but if you expand the lyric prompt that is not such a problem as style can be applied in lyrical prompt as well... (another possibility is to exclude tags from character length, meaning if you type [break] it would add those 7 characters back to the total of characters available for lyrics if that makes sense)
I have been saying that I want to use my own voice to a little ways up I had just posted that maybe yesterday or the day before, I do hope that they will consider letting users train Suno with their own voices if they wish, and then would need some way to recall the voice when you want to use it or maybe a way to set it as permanent, I wouldn't mind my voice being used either, but me being able to use it for what I am producing is exactly what I have been waiting for.
Hi again Suno.
This recent change to remove the ability to cover/remix other people's songs has essentially killed collaboration, and it wasn't even announced in the channel (which is quite frustrating). It should be the artist's choice whether their song is open to covers or not. What if I'm working with someone on a collaboration, and we want to experiment with different ideas through covers? Restricting us to the 'Extend' option limits creativity and it's making the platform less appealing for collaborative work. could we bring back the cover option, but with the ability for creators to disable it if they choose? Thank you.
Can you give us an option to create our own voice model that we can use for every song we create please and thank you
Flosstradamus's song is still cover-able which basically adds onto my point. Make cover's optional.
You can clearly do it, so. . .
Nah, It turns out that it worked better this way:
[Verse]
Song Lyrics
[Voice Type (Female or Male)]
I'll share some examples with you:
https://suno.com/song/59792aa8-3281-46e2-b42c-7bd88ed494d5
In this song, at 00:58, I'm supposed to have put "Female Voice" in the prompt and it turns out that it took the prompt from below to determine the type of voice, meaning it didn't take it from above the song lyrics but from below.
So to verify this I decided to do another experiment generating another song ( https://suno.com/song/bc7bdeac-4dcc-413a-9078-50d0fe8be5e6 ), but now adjusting the discovered format:
[Verse]
Song Lyrics
[Voice Type]
And it performed the command again as I wanted.
Pop Country, harmonica instrument, violin instrument, two vocals, male & female vocal, duet vocal, 120 BPM, song. Listen and make your own with Suno.
Country, harmonica instrument, violin instrument, two vocals, male & female vocal, duet vocal, 130 BPM, song. Listen and make your own with Suno.
Suggestion for song creation.
Sometimes the output it ALMOST perfect just needs a big drum section for the dramatic at just the right moment like the second verse to keep things interesting
Or a change up on the bass line and synth to make for a dynamic chorus section
Extending a song only does so much and I’ve wasted a lot of credits hoping it would add interesting elements but for some songs it just kept the same 4 bar loop even if it was perfect the song wasn’t dynamic enough and generating another creation with the prompt would prove to be unfruitful because it would make an entirely different song
So we need something in between like chaining voice samples or something so we can add more than 60 seconds so it has a better understanding of the song intended
Please put an option to have library search results order chronologically.
For thee looveeee of gooood, pleasseeeeeeeee increase the 60second restrictions on upload feature, pleaseeeeee
lol
make it like 2 minutes or something
I'm building a beat that is f*cking killer, but it goes to around 1:50, until I get stuck and i don't know how to proceed, so I'd like to upload it to see what Suno suggests, but I can't and when I cut off after 00:59s, then what Suno suggest after isn't comparable what I have created beyong the 1st minute
ActivityPub Integration
I'm a backend gentleman, of course I'm interested in protocols. I think Suno is well positioned to integrate into the fediverse which gives Suno and its users a lot of benefits. I recognize this might be a hard sell (And probably a long term thing), but given the team seems to be moving in the direction of adding more social components. Here's my pitch.
The first is your platform and your user's discoverability goes from basically zero (Suno itself) to the millions of users not currently on Suno, but that are using products like Threads, Mastodon, etc.
The second is to Suno directly, which is ActivityPub is an absolute firehose of data which could be used to improve the model.
The third is you don't need to invent a social API, the spec is well defined, and not that difficult to implement.
The fourth is there is no one at Suno's scale and quality dedicated to music on the fediverse at the moment.
Albums
I'd like to see albums. "Isn't that a playlist?" No, a playlist is a random mix of songs. An album is the packaging of a completed creative vision. An album would need to have a couple things that make it different than playlists to be cool.
- Liner notes
- Ability to add a handful of photos
- Ability to either upload videos, or link into youtube for music videos
- Ability to control continuity between tracks (track 2 should start -2 seconds into track 1)
- Ability to link into album pages on the streaming platforms.
- An easy way to remix an entire album
- An easy way to highlight other Suno people that collaborated with you on it.
can we please have an option for double veres or 16 bars for hip hop artist? The 8 bars are ok but there should be an option for extending the verse for 16 bars. Also a hook option that adds the ability for vocals i have a work around for it but the verse should be empty for the performer. also can we get higher quality downloads
Suggestions for suno:
Mascot. Personify the bot. I wanna know who I'm fighting with.
"Keep vocals/instrumentals" -Edit existing song lyrics/instrumentals, keeping one of these things, the same.
Song cut off point
Save/use samples from your own creations
Allow us to create bands.
Choose vocalists/instruments separately from prompt, attaching styles to each
Save vocalist/musician
Customize vocalist/musician
Add vocalists by name
Segmented lyrics boxes, with a menu of options for what segment of the song they represent. For instance, you can select intro for the first box, and actually get one of those, or if you don't want one, select for the song to start with the chorus, a verse, a hook, etc, and then, have the option to add the boxes you need.
Regenerate/repeat song section
A list of clear command [Tags], (Backing vocals), and $sound effects$.
Find extensions (Just a reverse of the "Extended from" link)
Trash search
Mass perma-trash
Trash folders, per song
Refund credits on bugged songs, and allow us to type in, and specify things, or add more specific options, that would allow us to explain, such as unwanted repetition of a verse, or mispronunciations.
A command to elongate words. ~ Does not work for me, and spelling it ooooout will have that word pronounced, as oot.
The ability to chat with the AI directly, and leave notes. "Hey Suno, this is pronounced like the tears you cry, not like rips in a piece of paper."
Maybe give the option to reject songs, and get your credits back, for paying customers only. Just make it so that, like... if the song is downloaded, it can't be refunded.
Also, I apologize if any of this is actually impossible. Admittedly, I have zero technical knowledge on any of this. You could tell me that AI, was really just three magic gnomes, that you keep in a box, and then, show me an AI generated video of the gnomes working, as proof... but I feel like things like these, have been done elsewhere, with other types of AI.
Thanks for suggestion. I'm trying Bandlab and if that wont work, I'll try "Too Lost". Appreciate much 👍
Also, since I write my lyrics and use my own melodies, I call it Hybrid Music Productions ⭐ not "AI produced music" because thats not entirely true.
Please add tibetan language,Trying to generate in it just comes out singing la la la. Its a minority language but it would separate Suno from all the other music generators. thanks
I've been using United Masters since Oct 2023 with Suno V1, never been rejected for the music, half the price of Bandlab's annual plan too. Bandlab is newer to distribution so they are a little more strict you may come to find out
The only time the devs have specifically suggested what Suno is, was a floating ball of ferromagnetic fluid
That waveformed to the music in its floating ball shape
i need paypal payment... i got no credit card T_T
Does Bandlab cost 15 USD? Too Lost is 3 USD for an Artist account and 6 USD for a Label account (where you can upload albums from different artists if you create music of different genres).
Just tested crop and edit lyrics on iPad and the interface is a bit odd but does work
An option to only generate one song instead of two is such a huge oversight, why hasn't this been a thing from the start?
The Crop tool should be moved from the context menu into its own menu on the left side, under "Tools" or just "Crop"
the opposite of crop should be a feature called Splice. A sound effect splice would be cool basically like audio inputs but instead of making a song based on a sound it lets you place specific sound effects in parts of a song you liked. Would be cool for some hip hop rap could put in owl hoots or other interesting ambience effects.
the new Crops tool love it but also new Song Detail i don't like it shot off the audio and after go back to create page all i have on the left side is resets i don't like it
Suggestion for a Song Review Feature on Suno
I’d like to propose a Song Review System for the Suno website to encourage valuable feedback and enhance user interaction. This system would create more opportunities for users to receive constructive reviews on their music while helping others in the community, ultimately providing increased exposure for their work.
-
Open Song Reviews: All songs on the platform can be reviewed by any user at any time, fostering continuous feedback and engagement.
-
Credit System for Reviews: Users will earn credits for reviewing songs, incentivising participation. The more reviews a user gives, the more credits they accumulate.
-
Double Credits for Reciprocated Reviews: To further encourage engagement, a Song-for-Song feature could be introduced. When users leave a review on the original poster's (OP) song, they can attach one of their own songs for feedback. If the OP reciprocates by reviewing the attached song, the OP will receive double credits. This promotes mutual interaction and helps less-recognised users gain exposure.
-
Gamification: To encourage continuous participation, introduce competitive elements such as:
- Leaderboards: Featuring top reviewers to highlight active and valuable contributors.
- Badges & Achievements: Unlockable milestones for reviewing a certain number of songs or providing highly rated feedback.
Feedback Options:
- Quick Buttons: Reviewers can choose from quick-response buttons with keywords like Catchy, Great Vocals, Needs Refinement, or Unique Melody to give brief but insightful feedback.
- Detailed Review: Reviewers can leave in-depth feedback with suggestions for improvement, offering users a chance to understand their song’s strengths and weaknesses.
- Rating System: A simple 1-5 star rating system could provide a general score, giving users an at-a-glance overview of community reception.
TL;DR: A Song Review System where users can freely review songs and earn credits. Introduce a Song-for-Song feature for double credits on reciprocated reviews, along with leaderboards, badges, and quick feedback options to enhance user interaction and exposure!
Okay, one thing about the update I really dislike. On iPad whenever I swap to another window when I swap back it resets to the home page. It’s very difficult to work and do anything else, like chat.
Not even a debit (bank) card?
Intro Edit: I'd love to see a feature that allows users to edit the beginning of a song. Specifically, if a song starts abruptly, it would be great to have options for a fade-in or the ability to generate a short clip before the start of the track.
Extend Length: The ability to set the length of the extension. I often want it less than 2 minutes.
An audio to audio would be a wild feature like runways video to video but it remixes or warps the song to something else.
Uhhhhh you forgot something? We lost a major feature with this update, no way to do custom pictures
you could just use replicate in the mean time and finetune a flux LoRA dedicated to album covers should be simple could even always have the explicit stamp thing in the corner for each generation lol
You are on an iPad, right
Is it different on desktop?
Probably
It's in the big 'Edit' menu now
The same one you use for Trim
I am on desktop
I have this, no pictures
An idea: personal radio channels. The way I imagine this would work - each user has a public page that can be visited by anyone on the web. The page url has the user's name on the end so it's easy to 'switch' to another user. On the page, a user's songs play continuously, curated from a "radio" playlist configured by that user (or possibly the user has the ability to curate multiple "radio" playlists). I imagine this wouldn't be too much of a stretch to implement given that playlists already exist and there's a music player already in place.
Yeah looks like it's just missing
Aha, if I turn it sideways it’s visible, it’s there just buried
The message at the buttom currently states 120 seconds, but it should clarify that it's actually only 60 seconds for free users. and 120 for pro. You can only see it when you upload an audio, but it should say it from the start.
So my suggestions is simple: Clarify it before.
can we please have an option for double veres or 16 bars for hip hop artist? The 8 bars are ok but there should be an option for extending the verse for 16 bars. Also a hook option that adds the ability for vocals i have a work around for it but the verse should be empty for the performer. also can we get higher quality downloads
come on Suno, Crop should be enabled also for free users (I am a paying one), because unfinished songs are not songs !
It is like giving people unfinished presents !
I agree with you. The whole thing with distrubitors is a pipeline where people need to go the extra mile instead of just uploading to "Spotify". 15 a month is a grate deal if nothing finds it way back. I read some bad reviews about Too Lost and I already got 3 approved songs with Amuse, but they're closing my account. What happens to my music then? We're kind of in the claws of these companies. 🤷
A feature to set the TEMPO and KEY of a song when creating it
I agree, it's a basic feature and should be available to everyone. It's not like it's an advanced feature like covers or get stems.
Especially because some users, who would be willing to get Pro, currently don't even have the option to get Pro due to the limited payment methods available.
please can you make an option to use only same voice in our songs?
without every time with a different voice
I, on the contrary, read a lot of good reviews about Too Lost before I subscribed. You can also choose DistroKid. A bit more expensive, but they're not against the AI music, either.
Ok I'm putting my foot down for this one. Seriously. This is the SINGLE thing that absolutely makes me RAGE at Suno.
Verses trying to use lyrics from other verses. Is there a way we can make it strictly use the words we chose in that specific verse? I've dropped hundreds of credits on one song just to get it to use the correct lyrics.
Ola bom dia a TDS amigos(@$)
Suno team - you are amazing! Please get the vocals clearer.
Thank you.
Yeah an option to get (or generate) the vocals without the music would be a great improvement.
there is already: try "sans instruments"
well, Suno "devs": it was too difficult to copy somewhere else not only a CROP/CUT function, but also a FADING OUT function ????
To difficult in 2024 to copy/paste a stupid FADING, isn't it ? Even CHATGPT can program that in 10 seconds. What should we do with the songs that cannot be cropped cause they still sounds bad or you cannot have a sudden finish? Now we have to wait X more months or weeks to have also a fading out function? 🙂
I heard if you write '[strictly follow the lyrics]', it does it more often than if you don't.
I will give that a shot
SEPARATE PLEASE EDIT LYRICS/TITLE from the CROP function !!: it was so fast before the CROP introduction to edit the title! Now it is slow and takes a lot of time! Something we have to do very often now it is unbearable. Title and Lyrics we have to change very often, but CROPING seldom. What a futile data transfer and displaying graphics in vain, when you simply want to only change the title! I dont know who programmed and approved this, but was a very bad idea for all: bandwidth, time, mood, suno, users. Not to talk about the extra submenu of the menu of the menu...now it takes also one more click to dive into the edit text menu. What a pity... good function put in the wrong place !
Two very important points that should definitely be changed!
Point 1:
Please introduce a mass/bulk download.
This would be super helpful and would save a lot of time in the workflow. This function would be incredibly helpful.
Point 2:
Please please please, Separate the ‘EDIT LYRICS/TITLE ’ from the CROP function. The workflow after the new update is a disaster and costs so much extra time. Just to change the title alone. Maybe it's not so noticeable if you only have to change a few titles, but if you change many titles afterwards, it doubles the working time.
I am 98% happy with SUMO, if these two points could be introduced I would be 100% happy and satisfied.
Thank you
GENERATING SONG WITH THE SAME NAME ->CHAOS: Suno, it is difficult to add a counter to the names of the songs generated, as some other music AI does, so we KNOW which song is it? I have tens and hundred of songs with the same name. This is very stupid and changing the names takes a LOT of time!
@modest frigate team is aware, some of us guides and mods already reported the same 😄
you, me, the devs, all us know this from end of April... and now end of September, 5 months later nothing changed! This is one hour of programming !
At the most. Actually is a 10 minute thing! We dont write here for Suno to be "aware of", but to be implemented.
Thank you I'll try!
You need to add a function to the crop feature that lets you detach the resulting clip from its original chain. If I crop out the start of a part 2 to start a new song from that bit, it kinda screws everything up if future 'get whole song' sets everything prior to that.
Uhm.. crop is here since yesterday. So we have a new edit. It did work better and different before 😛
Please not that story again. They are working hard.
you did not comment that one, you commented this: GENERATING SONG WITH THE SAME NAME ->CHAOS: Suno, it is difficult to add a counter to the
Ah.. my bad
This does not work, how to do that exactly?
to work hard and to work smart is a BIG difference! The crop function "destroyed" the fast "edit title" ...
it worked for me...
Let's not start this Grabby. You cannot watch behind the scenes. There are more things to tend to and priorities are not clear to us.
ha ha, you want to tell a cook how much it takes to make an omelete :)))
I see already how much it takes to make some needed channels here! 🙂
Omlettes are simple.. a multi million dollar company is a whole different ball game
adding a counter to a file and songe name it takes less than to make an omelete... you forget that I am a programmer also,...and I was also product manager... I had a team of hw & sw helping me ...I know how much things take. All that changes that I requested so far I could have done with 1-2 devs in 7 days (plus testing extra) !
I didn't forget. But you have a tunnel vision. You cannot see what more they have on their plate, and who decides what goes first.
Did you already applied to their website?
yeah, defend them 🙂 they deserve...
Mabye you can help from the inside
I would have done this even for free 🙂
Since you have so much experience, you should just apply.
even you offering me jobs when criticizing the facts 🙂
I wrote them, but NO answer !
Because you say you know how to do it. I mean, I am sure you are running your own company.
of course I do
but what I am talking about are only UI simple basic stuff even every student could do... it is not hitech audio or ai stuff
If you are running a company like Suno, in size and numbers, you know it's not only about UI
wehn you are gonna be a programmer in the next life, you will understand me... 🙂
AI will learn me 😄
well, in a company with MILLIONS budget and hitech the UI guy does not have to do other things but UI!
only in 1 man show, the guy has to do all...
Eitherway.. No matter what you feel or think, feedback is alway helpful. But you have a tendency to bash the devs. And we have warned for that serveral times. Refrain from that.. 🙏
nobody gave more feedback than me...
but knowing how much it takes to programm things and seeing how much the Suno takes for basic UI things drives me crazy!
Not sure I can confirm that. But you catch more flies with honey than vinegar.
no, this wont work here... actually I have the impression we all write in a BLACK HOLE. Cause no feedback from devs , managers at all..
Only you are now and then contradicting common sense 🙂 But I still love you
You are something.. just don't know what yet 😛
I like suno too much not to want it to be better... 🙂
We are on the same page on that.. but sometims we have to learn patience
tell a cook an omelete takes 5 months to be made... 🙂
if we have to learn "patience", then what Suno have to learn? Wanna know ? PROGRAMMING (UI) ! 🙂
Meh... AI will do that in a few years.. maybe time for you to learn another job 😛
AI is doing that already! I have 5 jobs dear, very different ones.. I am not coding anymore. That time is gone.
so I am flexible...
Good! 😉
but nobody is flexible enough for the times to come... there are difficult years ahead of us...
I would like to add an addition to the cropping feature (epic btw, very good job well done). I would like the ability to re-order the parts, or even add a part from another song into the mix of this song. This way if I go back and edit later, I can make sure the edits are in the correct order.
I would also like to have an "inverse crop" where instead of selecting what you want to keep, you select what you want to get rid of.
@hybrid crane I have no idea how this DJ guy gave more feedback than the other 400,000 people here, considering his account alone was only made in june
Apart from the comedy club moment, I do have a suggestion. It would be cool to be able to upload more than 60 seconds of audio. I was thinking about the Soundclash remix contest and I noticed that it was indeed possible for longer content to be uploaded to the platform.
Do you think it is a server storage issue?
create some remix
If you’re paid you can do 120 secs now
We should also have the option to change sytles when editing a song afterwards 🙏
Especially for uploaded audios.
Really would like block downloading.
I think they should limit it by file size and not duration (In case it's due to server costs)
Maybe free users like 3-5mb and paid users twice as much.
Thanks SUNO for 120 sec upload now, this is much better, though not enough to make a cover of a song you have already. But it would be better to find this out from a notification in the app and not from discord, randomly.
I’d like to propose an enhancement to the current thumbs up/down system. While the thumbs up is great for marking songs we like, it doesn’t help users identify which specific versions of their songs they prefer the most. To address this, I suggest adding a dropdown box (with numbers 1-10) to the left of the thumbs up button.
This feature would allow users to rank their songs in order of preference. It acts as a simple yet effective tool for organizing different versions of a song, helping users keep track of what they liked best. By selecting a number, users can easily prioritize their favorite versions, which will provide a clearer overview and prevent songs from getting lost in the creative process.
The thumbs up feature is great, but adding this ranking system will help users make more informed decisions and stay organized, ensuring they can focus on improving the versions that resonate the most with them.
Nah dont have cvc on my Card 💀
Not sure if it has been suggested before, but how about upgrading the music player a little? I made some playlists and when listening to them it would be nice to have the possibility to set it to shuffle songs so I don't always have the same order of the songs 
I think the contests are a good idea, but they get too little impact outside of the Suno community. To get a wider impact, contents could be created in other platforms. For instance, a contest in YouTube, where the users should post their songs in YouTube and state clearly in the comments that they have been made in Suno.
I also think that not only song creation is to be promoted. For instance, creating playlist with Suno-made songs may be an excellent showcase of the capabilities of Suno.
For instance, creating a contest like "Create your favourite [jazz/dance/chill/pop] playlist" would be nice to get general people to hear music made in Suno.
Another option to get wider impact, it would be creating ambient music channels for business. Curating just four o five different playlists with about hundred songs (in no time with Suno ;-)) would make many business broadcast the Suno creations, since many business are eager to find free music but quality.
I'd like to suggest the ability to view all of your uploaded audio files, because apparently there isn't an option to do this.
There isn't, so since upload has been a thing I immediately put them in a playlist called "Uploads"
To your point though we shouldn't have to do that
Hi everyone. I am blind person. I'm using suno nearly every day, Unfortunately, when suno's interface was changed, working with screenreaders was little bit broken. It is possible, please correct this wrong.
if you can give us some specific ways in which it's broken for you, that would be really helpful for us. thanks, feel free to DM
Please make it so permission to copy your work in any way is off by default for other users. have a choice to never allow, case by case, or never. Having it where everyone gets permission to copy your work if published discourages artists from releasing their work on this platform. It is our work we should have a choice other than heard or not be heard. I would release more stuff but I dont want to give everyone rights to it so the bulk of my catalog is private. there is simply no incentive for me to share. thanks
its probably been mentioned before but
- individuial stems for each instrument used as well as vocals
- layers for the songs you extend or reuse. if I have a song called "Bobs yer uncle" and I make new versions or extend on it, it would be great if they were nested under the main file and I could close or open that group so see or hide the new files from the original.
- MP3 audio selection. a LOT of the time the wav files and MP3 files sound very low resolution. it would be great for subscription members to be able to get lossless audio or close to it so we can master it in our DAW setups.
allow pro users to upload audio up to 6 minutes long.
it seems the more i try to emphasize the gender of the singer; it wants to do the opposite. i wish i could fix it better.
How about instead of blocking files over 60 seconds (or 120 for pro/premier users) you allow them but auto cut them to the first 60/120 seconds (So we don't have to cut them ourself before we upload them)
Would save some time 🙏
Feature suggestion: for the song lyric generation, to have a parameter setting for "weirdness". Compare to Midjourney, that has a 'weirdness' setting. Less weird would use more conventional language, more weirdness would introduce more unusual phrases and words in the lyrics.
this would be so naturally and great, I proposed the same
Hello.
Please bring back covering and extending at the same time. Thank you ❤️
Is there any particular reason we can't use Cover on stems? I'd think that would be a perfect way to try and get an inpainting effect with current tools
Would it be possible to add a "Make Loop" option to finished tracks, that would edit the beginning and end so that it becomes a seamless loop? Like how some image gen AI has an option to make tiling images?
Hi!
I miss so much a "Describe" feature.
Ideally, it would describe any music audio uploaded.
However, it would also be nice to have it enabled for the songs generated for Suno.
For instance, when Suno creates a song for a simple "pop" music style prompt, it uses (explicitly or implicitly) other style tags.
It would be nice to get to know those tags in order to create more songs with a similar feeling/music style.
In fact, getting to know which style tags are useful and which are not is a major challenge in Suno.
I mean, you can use the tag "Christmas song" or "Noel", but Suno may not take them into account at all.
Maybe a "Create loop" check box would be nice when creating songs, for instance, for making background music for games.
That works too. I'd prefer to have it as an edit, so that if I'm extending tracks, I can still add the loop once I'm done. But anything that gets me looping tracks is awesome.
We need more than 3000 characters especially when working on covers.
QUESTION: Have anybody tried to merge in a daw a song from two different covers (version of the same song) and then feed the new created "song" via Upload to Suno and let it make a new cover of it? (there are some good solos or choruses in different songs created that you might want to have together in the same song to have the perfect version): I think this might work, even the merge point is not perfect? It would good to know if Suno can overpass even not perfect merge points in the song.
You mean cut and paste together a franken clip composed of the best sections of various bits and hope that the cover feature combined with the lyrics and style prompts can assemble it into something coherent and not needing further editing?
I think that the recent crop editor is a step in that direction. The next step could be auto-split the track into the [intro], [verse], [chorus].. sections to let the user reorder them, delete or, as you suggest, replace by sections from another song (given that the BPM is the same), I guess...
Could we get a filter in the library for Full Songs?
Would it be possible to add a option to translate or change the lyrics of a song for free? I am from a tv series community who likes translate songs, but unfortunately the site that has it is paid. Please can you make something?
Good afternoon, I am a pro user of suno. I am very interested in a couple of things. The first is that suno can have an editor like soundtrap where we can edit the voices and music in more detail. An integration like the one I mentioned would be amazing. The second thing I would like is to be able to add own voices or voices already generated on other platforms so that they can sing our songs and that it is not random as it is now. I think that these two features would make suno better and better and better.
The model is still largely not following directions for the structure, at least when there are no lyrics. Getting it to start soft and then build up to the main theme has proved almost impossible, and when it does, I'm sure it is a pure coincidence.
And we need the possibility to inpaint and remix parts of the songs.
Some type of organizational system (like folders) would be super duper helpful. I beg of you 🙏
we need folders ! Udio is implementing already the folders...
I would like a place where we can write a note on each generation - So I can write some thoughts about each generation which will help me choose which one to use.
We need some BASIC controls...
-
ADJUST VOLUME - All of the MP3s Suno generate are VERY quiet. I have to max out the volume to make them listenable--so they are NOT share without an external MP3 editor.
-
GENERATE INTRO - I have some wonderful generations with no into. Countless regenerations yield nothing with an intro comparable to what I have! Please let us generate intros to extend with the existing song.
-
PATCH A BAR or FIVE - I have several wonderful generations that are AMAZING but I can't use them because they contain (a) a few portions of lyrics done by Max Headroom, (b) chirps, (c) an odd instrumental for just a bar or so that I'd love regenerated.
-
CUT OUT A BAR or FIVE - Some generations are PERFECT and exciting but (a) contain repeated sections that make the song unnecessarily long.
-
EXTEND TO END - Some generations go ON and ON and ON and maybe hallucinate creepily after wonderfully generating, when they just need to END ALREADY. (Extending the end of a song with NOTHING but the text "[End]" or "{End}" is ignored. It's going to go on and on instead of bring the song to a quick close. Try it. I'll do 30 generations and get maybe 25 at 2:00 minutes, and the rest not much shorter.) I NEED to replace the end of a song with a short ending. I can get one maybe one generation in 40-80.
If you actually read this and you work for Suno, please reply. Thx! (Hey, Suno IS amazing. I just want to USE what I create.)
Hello.
"Style of Music" is entered on the HTML page in the right column with an additional and unnecessary space. It is enough to clean the string after downloading it from the database. It would be a good idea to clean it also before saving it to the database from all space characters at the beginning and end of the string.
I don't mean to be that guy, but can we please have an age toggle or something so people above the age of 18 can make songs with bad word? I'm not going for a cocomelon album here I want something me and the boys can laugh at.
Hello - I would love the ability to be able to edit a portion of a song. Sometime the AI gets weird in portions of a song, and it would be awesome if we could regenerate that specific portion of the song without losing the rest.
Yet again I struggle over and over and over and over AND OVER again with suno re-using lyrics in other verses for the chorus and for the next verse. This is ludicrous. THERE NEEDS TO BE A WAY TO STOP THIS!!!!
The cover feature needs a refresh on free uses each month until it's more reliable. It's like more than half of them ignore whatever you added to the prompt and just regenerate the literal exact same sound file. And the concept of using it to change lyrics like the popup described is impossible.
And maybe a drop-down menu on voices to select from so it can stop ignoring whatever you put in the style prompt.
And DEFINITELY the option to change tags after generation because often times what comes out is not the same style you plugged in, but the result is still good and worth sharing.
18+ version of the website, use any cover art and unrestricted lyrics
I've already accidentally gotten bare breasts in the cover art through the AI interpreting my lyrics, so maybe something can be done about that
That's an entirely different discussion unrelated to the terms of service vs. what the AI is doing
Average p**n supporter
ANOTHER THING: Updating the cover art seems to have been impossible for the last few days at least
Technically,
violate any applicable local, state, national, or international law, or any regulations having the force of law;
Because exposing certain images to minors (who have ready access to the site) does in fact violate some laws, and I'd assume literary content (incl. song lyrics) would be controlled in some regions Suno is available in too but I don't know for certain
i think showing naked breasts to a minor who didn't click "yes i'm 18" is quite a bit different from having a scoop of mint chocolate chip making your spare change sticky
anyway this discussion is pedantic and veering off the topic of the channel so back to suggestions:
In addition to "cover" there should be a "clean up" function so I can go back and get my old beloved gens without the random swelling white noise in the background that gets so loud you can't hear the lyrics by the end of the song
I'd love to see who likes my songs. There are like 5 people who seem to listen/like to everything, in my head it's the same 5 people on every song, and I want to form a club.
CRITICS: I am here since 4-5 months, but I have never seen a developer or project manager giving a SINGLE comment about bug fixing or requested features. Nor have I heard or read about a status page like Udio has, a page to see the status of the request is shown (in progress, in evaluation etc). We have no idea what all our suggestions are for and what happened to them. Nor the mods sorted out all the suggestions to put them in a separate channel. So people are kind of wasting their time writing things nobody cares... nobody sorts, nobody gather...nobody implement (with few excpetions you can count on one hand) and if somebody cares it keep silent as a fish. In a digital world.
It seems that the Suno team is rather hiding from their users and keep silent all the time. A kind of digital black hole this server!
In the User interface and song management, which are the most easy and fast to implement were very few and little changes within 5 months. In 5 months other make a complete new application (I am not talking about AI models, only bout user interface). I dont' know what happens there at Suno, but it is really weird to wait months for minor changer others do in hours, wait in vain for a word, a status, for a feedback.
Even the management of this server is very neglected, which is actually such an easy and fast thing to do. Cause there are not enough channels for all the different subjects.
Contests with SUNO songs
I proposed yesterday to allow SUNO songs in the Music League App, but I get 7 downvotes 😅 https://discord.com/channels/769318876885811221/1289273506537668660
Maybe the right place to propose it is here! Music League is an App to create Song Contests with your friends, each round has a topic and your friends submit songs, then you and your friends vote for the bests songs, and there is a classification. The songs must be in Spotify and the result is a Spotify list. More info: https://www.theverge.com/2024/4/18/24134059/music-league-spotify-algorithm-antidote
I was playing with my friends and family during the pandemic and was very funny.
https://emtv.cat/show/FestivalEmtevisio/Estiu2021/g2128-raffaella-carra-una-di-noi.html
I think that would be great to do the same kind of contests but with your own songs created in Suno
No one is forcing you to use this server.
After cropping a song, I wish it would stay in the edit screen so that I could then go and edit the title, lyrics etc...
no one is begging you to comment my comments..you are not Suno 🙂
Playlist Change Notifications / Playlist as a radio
If you think of a playlist like a radio station, similar to how DJ @potent folio and I'm sure many others use it, it would be cool if when you like a playlist you got notifications when it was updated (within reason maybe playlists you only notify you once a day to prevent people abusing this feature). Perhaps you could even create a radio style playlist that forces people to only add songs that they themselves didn't create to get people to help build the community. 🙂
All you do is complain constantly, about everything. Complain about the mods, the devs, the discord. Talk about how much better udio is. There is a very simple solution to all of your problems, one you can do at anytime.
You can on iOS, and there is a non-public API, so I suspect this is coming to Desktop. I actually had this exact same thought in my head it's the same 5 people on every song but what I found really surprised me.
In defense of the Suno team, if they responded to every single comment, they'd not be able to do their day jobs (acting on the comments being made). I've had them reach out to me for some items, so I know they are actively reading this channel. 🙂
Feature request: I would love the ability to download entire playlists of my own generations as a zipped folder of wavs, oggs, or mp3s. This would save me the trouble of manually downloading--say--my entire Drum and Bass playlist of over 60 songs, if I've decided I might want to use them for an impromptu Twitch DJ session or something.
Fine for you, but did this it happen privately? Cause I did not see ANY response in Discord and I am pretty active ! Nobody said they have to respond often.. but nothing at all?
Well said.
Privately yeah, but I see them occasionally respond in general as well after new features drop. There was a guy whose name I forgot spoke to them about adding Telugu as a language in general and they did shortly thereafter.
Ok, man! you think an announcement of a new feature is what I was talking about ??? God!
May a day soon arrive that fills you with joy, and when it does, I hope this difficult day fades from your memory.
well, you don't make a good attorney bro ...better leave it and find some other backdoors in Suno's site
I've gone as far as 10 minutes and 30ish seconds with Suno. I think others have gone well beyond that point. I only stopped because I was like, "this is getting too pretentious even for me." For example https://suno.com/@taboovector_music goes over that 4 minute limit quiet often it would appear.
I rembervrhis
I dont know of developers read here. But… The cover feature is very good. But is you upload a 2 minute song of mine world be very Nice if suno could have The option to save The instrumental without vocal. Sometimes it is a very good base with a stupid singer that perverts The original sense of melody of The Music that was uploaded. I with I could use The musical idea with My own voice. In this way of work stemming Is not an option since it degrades The instrumental Portion of The song when it is divide.
Would be not world be
Assuming the Filter has detected vocals during the upload, just give me a warning before going back to my library. Going through the process of selecting the right cover image, naming and potential lyrics and then realising it will not be public, feels like a waste of time. During that time I could've adjusted the clip through some extra filtering, perhaps choose something else and retry.
Before you say that I can do the naming and cover afterwards, sure. However within the upload section of the process, you can immediatly drag on a new file. So to streamline that process, it would be neat to get a warning that it will be a private soundtrack, rather than what I said previously.
Also, 00:00 extends pls.
Did someone already suggested an radio mode? With radio imaging SFX and stuff.
brazilian chat, japanese chat, spanish chat
Hello, I'd have a suggestion, it would be really useful to have the option to choose the language you want your song to be. If you make short jingles, sometimes there aren't enough words for the ai to pick up the language and most of the times it goes by default to English
there is a workaround for that: save the stems, take the instrumental version and feed it via Upload again in Suno and use cover... depending of the momentan mood of the interpreter you can get the same new version without a voice. Choose of course instrumental version in the prompt window
The stems degrade and flanged the instrumental tracks specially the attacks of instruments like acoustic guitars and pianos .i use to “stemming” in Logic Pro. I was covering my own songs I did several years ago. I up a part of the songs and guide Suno about what is Verse/ Chorus/ Bridge etc… It responds very well .Suno , in some cases, do nice arrangements but mostly the vocals are very bad. I was wondering that someday , in a future upgrade , the software could do ex: when you like the arrangement , have an option [without vocal] to make the same arrangement without the vocal and without the flanged results of stem. Just the clean arrangement. I will try to make your suggestion to see if Suno learn that I want the same arrangement, but as long as I tested similar situations this is a bit unpredictable. Anyway , thanks very much to spend your time in answer my question. Have a nice weekend. Regards!
If Suno cannot do that, maybe Udio can do that. It can import also songs and it makes much better covers, more predictable and setable. After that you can reimport the cover in Suno and you get the same sound
I’m guessing this has been suggested before but the instrument tracks being generated for higher tier users is something I’d pay for.
loop mode, specifying the end-ish of the track and having it blend back into the front and render such that it'd play perfectly
Feature suggestion: Could you add a feature that allows us to control what type of voice we want. I've had a ton of situations where the melody and instrumentals are perfect but the voices are off by a mile (Not that I know much about music.) Something that would look like this.
- Lyrics
- Genres or type of music
- Type of voice
- Title
Can that be implemented?
"I haven't seen a single dev comment about requested features." "You think an announcement of a new feature is what I was talking about ???" I would have, yeah. Grabby, what were you demanding, if a dev announcing a feature that people were requesting isn't the kind of comment about requested features that you wanted?
text color is hard to read here
Suno Cover seriously needs to improve to adhere to the melody of the input. like 7 out of 10 generations I made changed my melody to a totally different song. instead of just changing the genre and voice but stick to the same melody and chords. 🤦
i hope for the day when a subtitle file or xml of the lyrics can be exported for video editors!
also why cant the mp3 export attach the art and lyrics to the mp3's id4 data!?
Hi, now sure if this is the correct location, but I think we should really push for a "batch download" option.
Nothing more annyoying (for me) to finish a session of creation, and having to download one song at a time.
It would be very useful to be able to see what playlists a song is in on the song details page and remove it from playlists (that you own).
I just listened to many of the top Suno songs. All the vocals for them are distorted. I thought it was just mine but that is the norm on Suno.
Please add a feature that ensures the vocals are clear and undistorted.
Thank you.
Yessss
I really dont use The voices suno generate. But usually, when The song is in Portuguese it does infectious and ridiculous sertanejos voices and when are female voices something similar to Anita. In fact I am not really into these new contemporary Brazilian styles but our musical culture is richer than this. Suno learns how to do these sertanejo shit but don’t learn how to do frevo, baião, marcha de carnival,xote and other traditional styles yet.Even when you up a piece of these “unusual” styles it transforms them in sertanejo. I believe this occurs with other traditional styles of other countries too. So, let feed the monster and then it will have better references when we further try to do world and ethnic styles .
WHAT AN ANNOYING AND MESSY FEATURE to name all created tracks identically! This creats CHAOS and TIME WASTING!
At least a counter should be added there as an option (eventually in preferences) so any song has an unique name. And the "twins" that are generated in the same time should have something like "(Ver. 1)" and "(Ver 2)" added to the end of the song name
I think there could be a "fuse" option. Very often I like multiple generations, but I can only expand on one. It'd be cool to use one generation as an extension of another.
Not that its a huge deal for some, but for someone like me who likes to listen and generate as I go, to see what needs changed during an extension. I find it REALLY annoying that every time I click a track generated or an extension. My previously deleted "Styles" section that I clear to make more precise edits now AUTO POPULATES the previous styles chosen.
While normally not an issue, but whne you are looking for very specific changes and ALL the styles pop back up. It locks the song to very little changes as its it trying to follow the structure of the style as well.
For someone who in generating 2-3 times an attempt only to realize the changes will not happen because my 17 different style prompts to get the initial beat decided to return. I end up wasting ALOT of credits. Please, remove this. If I pull the styles. I would like it to STAY that way
Subway Stops On The Mass Transit System Of The Creative Mind
One idea I've been thinking about for awhile is this idea of how Suno could assist in creative inspiration as a series of subway stops. A stop represents one song by one artist in a particular genre. I'd have it play just the chorus of a track.
Did you like you this stop? Cool, let's keep moving down the track. Oh, you want to get off and create from here? You get off the line and you're prompt is like: rock, fuzz guitar, female vocals, under-produced, and it's based on where you got off along the way.
Maybe you don't like the line you are on at all, so you transfer, and process of traveling along down the line continues. This whole process does a couple things: it helps new users understand the system, it helps songs get exposure in bite sized chunks, it's art in itself.
You could of course do this exact concept as sort of swipe left / swipe right and that would probably make more sense as a product, but you know, where's the fun in imagining Suno Tinder?
SEARCHING/FILTER MASK in the Create Window: it would be great to have such a searching mask ALSO in the create mode, not only in the library.
Because we have to go in library to find some older songs made during a long day when we want to extend /expand something recently done which has no place anymore in the create queue. Going in the library means new window or loosing the content of the text field, time lost for sure. Udio has this function also in the create mode and it is a very useful thing. I was missing it since the beginning actually. But if you ever implement (I doubt that) in this case it would be better to have it at the BOTTOM (near play banner) and not at the TOP !
A LIFO list with 10 items for the lyrics prompts and for the style prompts is a must. This is a job of few hours, at the most, and it saves us a lost of time typing and thinking!
a quick idea, now that we have Covers (and they work amazing!), introduce a simple midi/sheet editor (or even just a piano keyboard) that would allow converting midi to creations (by first converting to mp3 under the hood, unseen to the user)
Suggestion: give us the ability to delay the time a song is made Public
Why is this important:
No one is online when I post songs.
So my songs don't get any likes until it reaches EMEA timezones.
I live in Americas.
Without this I either:
Have to stay up really late: 12:00 am (Midnight) to post songs
Or Extend yesterday's good song with v3 just so I can post it at 9:00 am
For the love of music, let us filter by "Instrumental". We've got 99 languages but "none" ain't one. This would also be extremely useful in the Explore and Search tabs.
Denoise feature for songs
When I extend songs much beyond 4 minutes on Suno they devolve into a noisy mess
Also, having a add voice over and existing instrumental without extending part of the song (because then I’m beholden to Suno’s interpretation of the musicality)
Stem to Song should be next now that Song to Stem is a thing adds more control to the generative process to those that choose that route
Suggestion: Upload our own voice sample to use as the AI voice in songs.
you can already do that
The "Upload my audio" function takes your song, but not your voice. It changes the song into its own
I'd like to be able to fix a typo (change 1 word) in a song, and have it stay the same song as before, with just that one word fixed. 😥
Export song to sheet music so someone can perform it
There should be a System Added called "Favorites" or a "Bookmarks" song section to better keep track of your Most Favorite Tunes that you're about to Un-Private , I already divide up my tracks by putting them into Different Unpublished Playlists , but by the end of a Month I have like 400+ songs in the Playlist , there should be an area for the Ultra-Favorites !!!
Another Suggestion , I Don't know how difficult this would be to implement , but to automatically sort by which Day it was made so when you go through Library Mode on the Pages you will have a Browser on the Left Where it will have the Month names & the Days of the Month that you created a particular Track , Then click on the Day and have every track you made on that day be it's own set of 20 songs per page Tracks !!
A simple little tweak: Extend from where you have paused the song.
Please, please, please, please...
Make your IA extract vocal as clear stem.
Please, please, please
Inpainting different vocal tones, i.e. changing male vocals to female vocals.
For the record, Suno wants the same thing. But it's tricky because the AI processes the song exactly as you receive it, combining music and vocals into a single file. Separating the stems is like working with any regular track—there's always some bleed between them. To get clean, isolated stems, they need to be generated separately, and so far, no one has figured out how to do that effectively.
I idea, is it possible during the creation process, to buffer the "vocals only" parts (or track as a DAW language) and merge it with the instrumental to have the final generation. And when the vocals are need alone, just export the buffer done before
As a moderator, I don’t have full visibility into everything they’ve tried, but I suggested something similar before, and they explained it doesn't work that way. Rest assured, though, they have some of the smartest people working on it. It’s just a matter of time. Like I said, they want this just as much as you. 🌼 💗

