#📚┃suno-school
1 messages · Page 192 of 1
india
it don't have all prices inside my head.. check suno.com for further informations
In europe the montly costs of the pro are 9 Euros
when you say commercial right for new songs made -- how do you define it? i put a lyrics and use suno features like voice and music and thus th final output is my own copyright creation under pro?
when you create a song in the free plan - then subscribe to pro and male a cover - the song is not for commercial use, coz it was made with free....
So, I subscribe to suno AI. When I use V5 voice for the last few days, it doesn't work for me. It won't let me generate. I'm writing this The creation failed due to an internal error. Please wait a few seconds and try again. I sent an email and tried to clear the application data and it won't work for me, but when I use v4.5 all, it generates a song for me, it started happening 5 days ago
v5 voice under pro?
please make sure to update the app
am using either laptop or the system .. not an app
hmm.. ok..
like it copy paste it... pro = v5
lets say i subscriibe to pro version and then i create a song usningall suno features.. either music or vocals.. so the output i can use as my own copyright??
ich dreh gleich am Pinsel.....
YES..
subscribe.. create.. do whatever you want with the song
all rights are on you
can i discontinue subscription ?
wel.. subscribe montly
and it is not automated billing every moth but one time right?
you have to cancel your submission.. otherwise it will be pay monthly
so for now it wil charge only 680?
suno is working like every service on this planet... subscribe and pay... if you don't wanna use this service anymore you have to unsubscribe.. otherwise the money will be charged every month
what i am asking that i go for pro plan so if i subscibe now.. will i be charged 680 INR or 8160? right away? how much i will be charged rigth away?
if you use the montly payment you can unsubscribe montly.. if you pay for a year then... well... you can unsubscribe november 2026
Check here
https://suno.com/pricing
thanks VRking.. 😄 😄 😄
so based on you guys input i will subs and with confirmation that during my subs whatever i create.. it comes under my own copyright
i have taken a whoe video screen shot aout the communication.. any legal isue comeslater.. i will produce it
thank you.. i will subscribe and test it now
correct...
thanks
Please help
So, I subscribe to suno AI. When I use V5 voice for the last few days, it doesn't work for me. It won't let me generate. I'm writing this The creation failed due to an internal error. Please wait a few seconds and try again. I sent an email and tried to clear the application data and it won't work for me, but when I use v4.5 all, it generates a song for me, it started happening 5 days ago
Is there any error when you tried? add more details ...what format did you use? mp3, wav? how long...etc..
Is it consistently failing?
Is the issue reproducable?
I hope that you include the steps to reproduce the issue in the email to support.
The rest of us have no issue with v5 generating.
I upgraded to premier, I think I can do wonders with Studio, but I need to some documentation and faqs on it, maybe even a tutorial, do those exist?
Check their channel https://discord.com/channels/1069381916492562582/1420792719969878028 also their tutorial on YouTube there’s a beginners guide on there
I don't know what to do or why it happened, it worked normally for a few days and then this happened.
Imate li kakav savjet kako mogu pokušati riješiti ovo?
Did you try to eliminate things that could cause the issue?
Like try a different lyric, is there a reference audio? remove it and see?... or try different style... or even try a different browser.
What about different mode, and try Cover, Extend... or did it just fail when creating a new song? is everything on v5 failing or just one thing...etc..
i tried and it won't let me add audio and i don't want to generate the song either auto text or my text
Do you have any advice on how I can try to resolve this?
Feel free to DM me and walk me through the steps and see if I can try it myself - would I get the same issue? send me the lyric too if you don't mind... let me try re-create the issue - maybe we can see some hint....
I'm not sure if there is a current issue on the server end, but every clip I generate is sped up for some reason.
For context, it's a cover I'm cleanup up, so it's also possible that it's simply the track that's somehow broken.
In Studio you can slow it down back to normal if you have access to Studio if you really like to keep that.
Maybe it's a glitch - it happened to me a few gens but usually I don't keep them and gen a new one... but if it consistently happened then something is wrong..
It's pretty difficult to hear if it's usable, and I do have Studio.
But it's a lot of credits for sped up and potentially slightly garbled clips - and also when I'm using Editor instead because Studio is too unreliable.
And too often seems to make the browser run slowly, so I'm not using it that much currently .. and many clips are just stuck in generation when using Studio.
With this issue of sped of clips, there must an underlying issue.
Good evening I'm trying to make a track with Absolutely No Sound, No Beat ,No Background how do I make it happen
The oposite happened to me the other day - it gave me over 7 min track with absolutely NOTHING.
But I don't think you can prompt for nothing.
Need help. I am trying to make a sad song. Sorrowful. Mourning. You know the type. The idea is the lead vocalist is female, and the chorus sung by a ghostly/ethereal choir. It does it, but the song still sounds far too "uplifting" despite my prompt telling to be soft, restrained, sorrowful. The female to be grieving, trembling, broken. And the choir to sound distant, ghostlike, and sorrow-heavy.
@half dragon thank you
I'm curious - why do you need silence track?
Good Evening.
If you need a track with NO Sound - then you literally have nothing.
If you need a track with absolutely no sound meaning you want complete silence, that makes me curious why I are you trying to generate a track with no sound?
The purpose of using Suno, is TO Make sound.
Mixing i have my lyrics
So you want vocals?
[silence] it will be hit or miss like meta tags be, but when it hits, or if you have silent audio already, clip it, copy it....for as long as you need the silence
For mixing and I have my lyrics
Lyrics and vocals aren't the same though.
Not sure exactly what you're trying to achieve.
Wouldn't it be same as leaving empty space on your timeline mixing with other stuff?
have the song paused for emptiness , then start up again?
I'm also confused 🙂
That's what I was thinking.
Most DAWs can create empty spaces.
If it's for mixing purposes, sounds likes the attempt is to mix in Suno.
Yeah - and you can record silent track in Studio too - plugin a mic, change the input type to mic on the track, then press record but don't make any sound.
Much easier and better to use Audacity, make the length of the track, silence it, save as mp3 or wav, then import.
..and it's also free.
@half dragon The problem I had with the song I'm cleaning up, I'm suspecting the reason for the sped up clips, could be because the track has stuttering - which is what I'm trying to fix, because the rest of the track is just as I want it. Not sure if this confirms my suspicion, but I tried to remaster it instead, and it also has the stuttering - so it seems it's a latent imprinting issue.
I’ve tried this but only by downloading the suno voice, removing all effects from the voice so it’s dry and raw, and then using my voice clone to sing over it
The issue is that I don’t know how to add back the voice effects like reverb etc
Should I not be removing the reverb and stuff before I put it into kits etc?
Do you have Suno Studio?
If so, you can use the song without the vocals and add your vocals.
This should be the way you do it, it should do a mix with reverb, delay and such.
- Basically, you'd go into Suno Studio and add in the stems from your song.
- Then add in your vocals from your clone voice.
- Select all stems in the track + you vocal track, drag them to the custom box.
- Then in the dropdown select "Add Vocals".
- Hit create.
Yes I’ve got studio.
I can’t sing which is why I want to use the suno vocals lol
And voice clone over it
Did you try to Cover it again? I had a track that has issue of skipping but I like it so I Cover it again and it fixed it
I've created several covers, but this is the first I got that is what I wanted.
I haven't created others after it, because of the variety it create before it.
Suno doesn't clone your voice - yeah your best bet is to train your vocal with kits.ai or any RVC trainers to create your voice model, then use it to replace Suno vocal track as if you were singing it
Yeah that’s what I’ve been doing
But I remove the reverb and other effects from the voice before cloning over
Or should I not remove the effects?
try it both ways and see
I assumed it would sound glitchy and bad if I didn’t remove the effects
Also only had one free trial
Guess I’ll pay and see
But if anyone knows, please do let me know
Try Replay - it's free and it's local - so you can also train with it if I can recalled the last time I used it a while back
https://www.musehub.com/app/replay
It also lets me replace the vocal track - try it and see if it's still like what I described.
Thanks I’ll have a look
Is it just a small section? or the entire clip?
Small section might be easier to replace in Studio or Editor but if it's entire clip then it'd be very difficult to fix
You mentioned you don't want to use Studio because it's slow? or costly? with just a speed change - it wouldn't be costly or anything if I can recall.
Load it in, change the speed of the clip, Export!
It's a few sections.
Basically... Intro runs fine, then closer to the end of verse 1 and running through the pre-chorus and chorus.
That's what I've been trying to do, fixing in in the editor, but the clips are most of the time sped up, by a lot.
It's not that it's too costly in Studio, but unreliable and slow, basically this happens:
- Either clips get stuck in generation.
- Browser becomes slow or unresponsive.
- Clips often sound so different it's unusable (even with weirdness and style sliders down to 0 and audio influence to between 90-100.)
..and a few other issues.
The problem is, I have no way of knowing if the sped up clip is actually usable, mind you it was made in the editor, so I would have to leave the editor to go into studio, to slow it down and possible actually add it in the Studio.
so i am having a problem when i give suno a request to do music breaks and what the music should be, eg
[Outro — 25s instrumental]
Soft piano sustains under faint breaths.
A final chord lingers, fading into air.
Two voices sigh — “we remain.”
It makes the voices sing this, how would i stop it?
If you're removing elements through a DAW., and have a dry voice, what to stop from then duplicating the vocal and adding a reverb on the duplicates. Duplicate and adjust a few settings on the reverb tracks, pan one leftist and one rightish. Upload and layer the tracks. Adjust volume of the reverb tracks as needed , likely about 15% down.
Because I have zero experience using daws
And would never be able to replicate the quality that was made in the suno voice
Im removing them through kits
Not a daw
Anyone else dealing with a studio bug where the AI either flat out ignores the prompt and does its own thing or else gets stuck generating something simple like "war drum undertones and cymbal climbs?" Not sure what's going on cause it was working fine last night...
I thought you were using DAW for the "removing elements " bit.... Kits.ai, i see. Got it. . But really, try audacity 's free program.
Upload the vocal file. Under tools or effects , choose reverb and let it add at will. Dont really have to pan, but panning tool should be just a slider to the left or right as well. Pretty easy stuff to figure out Id hope, so the suggestion is still doable.
Otherwise, you can even try to just duplicate and layer with slight differences to get chorus effect. You might like.
Have you at least uploaded and see if that dry vocal matches back to your original instrumental stem? We may be putting the cart before the horse if they dont match well after editing in kits.ai
Cool I’ll have a go tomorrow probably.
On your last point what do you mean?
The dry vocals matches the song yes
If that’s what you’re getting at
It just doesn’t have the reverb and echoes the suno voice does
Ytou took the vocal l stem downloaded ok. It matches. Good.
Yeah, but it just isn’t anywhere near as good because it’s just dry without reverb, echoes, layering etc
Put them in [...] - Suno will not try to sing them if they're in [..]
[Outro — 25s instrumental]
[Soft piano sustains under faint breaths.]
[A final chord lingers, fading into air.]
[Two voices sigh]
we remain.
They might not all work, especially [Two voices sigh] - but at least it won't sing it.
recommend putting a line break between them as well
[Outro..
[soft Paino...
[A final...
Thx
Hello guyz. do you have any tips how to make instrumental prompt longer ? Extend option is total mess an not working like should.
Try to use lyric box but don't add any lyric.. instead... add song structured tags ...and build it as long as you like
You can build it however you like...
Also you can try adding prompts in [...] too to guide the instrumental
For example, just expriment with different things... adding ore structure sections and give direction using prompts in [...]
[Intro]
...
...
[Build]
[Chorus]
...
[Instrumental Interlude]
...
[Guitar Solo]
...
...
[End]
That is good idea, was thinking about that . Thank you , i will try this way .
Can i promote my release in this group?
can someone help me with a prompt for this song?
https://open.spotify.com/track/5GmNq5Vxgqb43EI0scxzAY?si=PDBxYAdcSf2tnYAzw5WWNw
Extending Song Verses - can anyone share their workflow? I am struggling so hard. Imagine, I have a Verse, Chorus, then musical interlude generated. I want to add a Verse 2. Right now, I go to Reminx, Extend, I move the slider to where the extension is suppossed to start. Then on the lyrics, I put in the new lyrics. Generate.
This has only worked once for me and when it did it was perfect. It matched the music and seamlessly added the next verse. But since then I have tried 20 to 30 times with no luck or success. It won't use the lyrics I am adding. Or it will stretch and carry on and remix previous lyrics and repeat them, then will briefly use the new verse. Or I just get a mess of garbled lyrics.
What am I doing wrong? If you are extending should weirdness go all the way down? I usually just try match what originally made the song.
ehm guys i have too much white noise and too much wet effekts in the music
i mean a little bit is okay but sometimes its too much noise and too much wet
add things like "no reverb" to your style. or "dry"
What happened with suno>open editor? Once i open it on certain track, it starts to play random music from my library and i cant stop it😆 i saw other ppl have same issue
Sorry probably wrong channel😁
Is SUno acting weird if you use a VPN?
First whats your subscription level.
Then, What's your experience in prompting , so far, since youve been using Suno almost a year. Or rather, what style of prompting do you prefer using?
Most who ask this want the prompts to match exactly. The korg synth used, the 808 drum style, what type of emotive prompts to get a dark or bubbly or ominous cinematic sound, and so on.
Typically we start with a genre analyzer app like audio Ai dynamics, or sonarteller.
We can search some recent lists of suno prompts (they are everywhere) for a vibe, or emotional prompts that fit the sound.
Can even split the song into stems if we wish to pinpoint what instruments might be used.
Throw some prompts in suno, get a outputs with that same vibe. Download. Upload and let suno analyze it further, possibly giving specific instrument reference, tempo, feel, vibe.
Write it all down. Put back into suno or just cover the upload with some lyrics that also come across with the same feel of the music.
Cross fingers and hope youre close.
i got Premium plan.
I got decent experience, but sometimes the AI just bypasses it and makes up some random guitar stuff.
Yeah I want exactly kind of the same beat pattern/instruments/vocal flow.
maybe you should try using Studio instead and multiply your track. just make sure to stem before working, so you have much more control above instruments. But also never forget, that AI is still a blackbox, you will never get, what you want. You just have to work with what you get, or you make your own stuff and upload it via +Audio
by using Instrumental Prompts? But to make sure, what're doing you should talk with chatgpt about how instrument prompts working and are spelled
Is there a way to put emphasis on specific words in the lyrics prompt?
what kind of emphasis?
Sorry. Im working now. And cant get back to this at this moment.
contrastive metrical stress
Anyone?
You can share in showcase https://discord.com/channels/1069381916492562582/1083129341870878780 or the genres channel if it’s a Suno made song
Yo, I have a song done in Suno which I like and I don’t want to change this song at all but there’s one word I want to change but keep it so it sounds the exact same but just a different word. Is there a way to do this?
Using Editor, highlight the section of the word, put in the new world, then click Replace.
Hi I am a free plan user, and what I decide to do is upload my instrumental audio for Suno to generate the vocal to me, other than that, I have a little lyrics with me, I need Suno to extend the lyrics for me, and it may change the sequence of my lyrics and even slightly modify the lyrics as long as it remains the same theme. However, when I have done all the setup and prompts on the interface, and I press "Create". It shows that "An error occurred. A system error occurred." This error message is too less info for me to understand what should I do to continue my work. Is it possible because of I am a free plan user? By the way, I have 30 credits left for today. Is it possible that I have less credits to generate what I want? I am still new to the Suno and ya, hopes that someone can solve my problem.
maybe you using some reference which are not allowed
If you say like create a song in the style of XY, you won't get it
It might be your prompt, which is not allowed. Dunno what you have typed
whenever i open editor it just starts playing random suno songs i created lol
i cant turn it off
common glitch?
hit me up when you can :) thanks
ok im in editor, trying to change a specific word (i like the whole song but want to change one word)...
changing from "you're frameless" (original) to the new one "you're shameless") but it just keeps saying "you're frameless"
any ideas?
Anyone know if its possible to make covers of actual songs? It wont let me upload audio of a proper song, i just want to make a custom cover for personal use.
i dont think u can upload any copyrighted material so i assume no
Thats unfortunate

There no way for me to like agree to only have it for personal use if its copyrighted?
You can't.
I got that one too. Very annoying
Working on some acoustic songs and I am having trouble figuring out a prompt for the guitar that eliminates the squeak effect often heard when from someone's hand sliding along the fret board. Any advice or tips?
i kept refreshing and it stopped
its called finger squeak or string squeak, try to just put in the prompt "no finger squeak or string squeak sounds"
I will try that. I just did it for a few times.
Yeah and clicked on an off too from editor page
hi - what should i choose in order to use uploaded vocals and let Suno keep the same vocals and not change them, just build instrumental around them? tnx
There's no way to leave the uploaded vocals unchanged via Suno - it'll replicate them with an AI singer.
If you let it do that, and keep the audio influence high so that (hopefully) Suno creates instruments that fit to the vocals and cadence as they are - then extract the instrumental stems from the song created, and either;
a) If you're a Basic or Pro user, connect the instrumentals back to the original vocal recording in an external editor like Audacity.
b) If you're a Premier user, connect the instrumentals back to the original vocal audio upload via Suno Studio.
thanks. but maybe i'll try to be more specific, let's say i upload some animal effect, not singing.
and i want to make Suno "improvise" with it, build a melody around it, even if changing it to some AI. is that possible?
In studio you can add new tracks and use it as reference
I can't say it's something that I've tried, but if you keep audio influence high, weirdness low and style influence mid-to-low and are willing to go through a few repeat tries until you find the sweet spot, it should be possible to get something from it.
yea, i was afraid of that 😉 ,
i guess there's no escape from paying more.
i just hope it will be worth it, and can really do what i described..
Well, like Kanthric said it should be possible to get something out uploading it and covering it
of
Studio just gives you slightly more control of where it is etc
@tame shale You shouldn't need Premier level to experiment with this. If you really want the animal sounds to survive the process, you could also try creating a persona of the uploaded audio and use that as well as the audio itself when you Create/Cover - that gives Suno two of the same reference points, making it even more likely to influence the output.
Studio is only useful if you're looking to re-import the original audio back into the final version, which you won't know until you have a workable song anyway.
I'd strongly recommend experimenting before you upgrade, since this is stretching what Suno was designed to do, so it might be quite a challenging project.
yea i agree with that, so what you suggest makes perfect sense.
Good Evening Europe and good afternoon USA 🙂
Suno is not following directions today... drum solos are not played with guitars ... thats why they are called drums. 🙃
drum solos are not playing with guitars??? that's paradox 😄
i think you mean "drum solos are playing with guitars" ??? 😄
nope,,,,, I needed a really good drum solo,,,,, I get a really awful version of Whipping Post ,,,, I sat here like an idiot (expressing very loudly) where are the drums! the dog is quite startled lol
Heys guys, sorry to ask, but I cannot get rid off the "backround voice".
Almost all my songs has some kind of choir or echo in the backround, or a voice repeating last word/words of EVERY verse.
I used ,,exclude styles" and wrote this:
background “ahh” or “ooh” choirs; wordless choir pads; background singers; extra harmony voices; multiple lead vocalists; lead voice swaps; long vocal delays; echo repeats of lines; call-and-response copies of the main vocal
and yet, it is still doing it
Yeah.
Any idea how to prevent that?:D
if i would know that i would win the grammy award 😄
the best workaround for me is creating my songs with v4.5 or v.4 and create covers with v.5
but that also depends on the style
have you tried adding them into the negative prompts?
Yeah I did, tried multiple ways.
Didn't work sadly
This did tho
v5 is still in a beta, so it makes sence it has some issues in it
yes... 100% agree.... and at the end of the day it's AI
Is there a way to prompt the tags for an instrumental, so they show up in the editor?
I would use meta tags and sort them out if they had a lyrical format but with meta tags instead. I would describe it that way with ChatGPT and it would show me an example of one. https://suno.com/s/WG3QoVCXoHk9XNiJ
Uh, some kind of "trick" with the asterisks (*) ?
Oh okay I thought with brackets like that.
Asterisks are sound effects from past comments of that before
Oh, I always use those brackets. But additional asterisks did not help right now in 4 test song creations 🙁
There are no sound effects in your example song, at least not in the public lyrics display...
You can try and ask in general the users in there might know more of that
I’ve never had any luck with sound effects either
Oh, I thought this prompting-channel would be the right place, but may be we need more people power 😉
I did, but it's like gambling ^^
What is causing this and can I fix it myself? It says I'm not subscribed, but I am, I have a v5 voice and 5000 credits available, and when I try to generate a song with v5, it says this for days. Creation error The creation failed due to an internal error. Please wait a few seconds and try again.
You can try creating your own sound effects in studio as a base starter and experiment with that in your song
I know, but I did not try it yet in my latest songs since Studio is available. But I should keep it in mind.
Experiment with it and see what studio creates based on it. I’ve seen there are some at Pixabay for sound effects there too 😊
I often have some problems with covering and extending songs in v5. In most cases the songs mess up somewhere and repeats a lyric which messes up the whole song and another time the ai starts mumbling the lyrics at one point. Is there a way I can fix that? I used v4.5+ and it works fine but the instrumental in v5 is less empty which I prefer
Seems you kept asking the same question for days - questions that only support could help you and you have already sent email to them -
Just wait for their responds and please visit #archived-ticket-help if you want to follow up.
That'd be more effective than keep posting here.
I need either help or more practice but I for the life of me can not fix just like, 4 simple words in a song of mine without it going all to hell in the process. Is there a step by step to change just a few lyrics? I mean I KNOW it can be done but just seems like a ton , "for me", to do for just 4 words.
Hello everyone. I want to clone my voice and make it sing my songs in suno. Whats best third party AI for that?
I tried suno recorsing and persona but just doeant sound like my voice..
What can we do about this annoying "Copyrighted lyrics" error when we upload our own songs with our own lyrics?
Does anyone knows how to do an instrumental that Suno doesn't trying to push out as a song? I want to create a simple very quiet, slow paced instrumental of dark ambient, which should be...simple, very quiet and slow paced yep, throughout the whole 4 minutes of whatever, but despite countless instructions after 2 minutes Suno pushes a beat, and energetic chorus out of nowhere, without asking, there's nothing in the instructions that would suggest it and eh.
Does Suno have "creating songs" deep rooted and it's impossible to overpass it? The prompts I've used: Create a long, continuous, beatless dark ambient track. The music must stay slow, minimal, and tension-driven with no percussion, no rhythmic pulses, and no tempo. Use deep drones, low pads, subtle dissonance, and very slow tonal shifts. Integrate soft environmental textures that remain consistent throughout: faint wind, distant metallic resonance, low rumbling air, quiet creaking wood, light rustling, hollow room tone, and gentle friction noises. These sounds should blend smoothly with the ambient layers, never forming patterns or jumps. The overall atmosphere should feel unsettling, quiet, and suspenseful — “too still,” with a cold, subtle unease — while maintaining a uniform, continuous flow. <-- and a few other versions of it.
My end-goal is to have a track like one of these "calm music 3 hour playlist" that is very neutral, isn't a song at all, doesn't have a chorus or whatever, just continous track of slow paced calm music.
.
It should be flagging anything lyrical, melody , branded or trademarked that are already publicly available.
That assumes then, that they are NOT therefore exclusive unique creations by the user.
So thats a wide spectrum when a phrase like "come on down" can be flagged for already being found in other lyrics , for example. Reporting , I guess in bugs-web channel, has helped , i understand, to white list some of the phrases so the censor is not so harsh.
thanks bro for the reply, and ya, what I found out is, today I open this again, and i try to change the selection from "Cover" to "Add Vocals" after I add my instrumental audio, and ya, it immediately ask me to upgrade to Pro plan, so I guess there are something wrong with the system yesterday, and today it fixed. I haven't prompt for anything, and what i do is just simply upload my audio, and try to add vocals, not even adding lyrics or styles prompt and etc. So, I think nothing I can do with current free plan if I wish to finish what I wanna do D:
Are you a paid subscriber? Tried doing it all in 3.0, 3.5 or 4.0 then for this? Pardon the Spanish, Suno apparently taking a siesta.
not yet, for now I'm testing Suno with free tier (I wanna go Pro, but yeah so far results are not good, so it didn't give me a reason to go Pro). I haven't tried anything other than 4.5, but I care about the quality of the sound and the previous versions are worse so I'd like to avoid regressing it (because even if the Suno will understand the instructions, the track itself will be lower quality, and I don't have assurance this will work on v5 if it works only on older versions)
It may not take one shot. Might run originals in 3.5 then see if remaster in 4.5+ will remaster it. I havent tried that step as I loved 4.0 for that, but 4.0 now taken away. Im planning on writing more 3.5 songs for Christmas so testing coming soon, though.
Otherwise, [ambient] and [dreamy ] are always good to add to calm things down. Exclude as many genres that have drums and percussion.
In pro then we can just split out the drums and percussion. But basic is harder. The skills of Suno last year's overpowered the limits of 3.0 and 3.5, so now with pro tools, we dont need the struggle anymore. That alone is reason to use a higher tier.
Frankly, I use 5.0 more to cover or remaster, not for creation when I want more melodic pieces.
Can someone tell me why some video-downloads is about 15Mb and other songs that are shorter is about 50Mb?
For those who haven't used Suno Studio. Here's a brief example of how you can greatly improve backing vocals. First example being the stem from the original gen, 2nd example was 2 takes panned left and right.
I like every part of my song but the chorus, what's the best way to redo the chorus only?
Suno Studio and regenerate the chorus.
Generally not a good idea to include negative prompts in the main style box (e.g. no percussion, no rhythmic pulses, and no tempo) as Suno isn't very good at distinguishing those. It's also not trained on sound effects, so your steering there may be more hit and miss than usual.
I've not tried creating anything quite like you describe before, but Suno does like to 'songify' its output, so this might be a little beyond what the current models are intended to do. You could try extending from near the end of the section before Suno takes it into a beat using the same prompts, and see if that helps to keep it more on-track with what you're after.
yeah yeah I was told before to not include negatives forgot about this one 😄 doesn't change the outcome tho when I deleted them now. Hmm so in general from what you say it will songify each track either way :/ Do you know whether it's possible to somehow search the Suno library? I think searching for instrumental-only songs or being able to filter the calm ones it would be the quickest way to find good ways/prompts
that's what I do with MidJourney for generating images -> going through other people's prompts and outputs made me learn the most, would be awesome if it'd be possible on Suno (although only with filter option, because a random showcase page won't let me find what I'm looking for)
@royal plover please direct your how-to questions in this channel. Thanks.
Your original question ---> #archived-ticket-help message
Can someone give me some advice? How do I add an Instrumental between the Chorus and Verse? I been adding it like this '''[Instrumental: Groove-focused, playful electric piano, bass, and drums, Guitar licks and subtle synth textures]''' I keep geting lyrics in it.
When you say you're getting lyrics in it, is that bleeding over from the verse before or chorus after, or is Suno just pulling them out of thin air?
Suno is just adding them in.
There isn't a way to directly search the library - the indirect method is by using prompts to flag up keywords (which is what Suno essentially boils down narrative prompts into) that you're asking it to reference against. So the more music-relevant the keywords are, the more likely Suno will find examples of what you're after.
That's unfortunate...and annoying. And that's happening for you consistently across multiple songs?
My first time trying to do this in the song editor. I think I can live without adding it, but it would be nice to give the singer a break before the verse.
Might be better to split up. Seems its ignoring it and just filling in with singing.
Use some variation of these might help....
[Interlude] [groove-prominent]
[Piano solo] [jazz] (bright, playful)
[Interlude] [guitar]
[Synthesizer] [free interpretation]
[Interlude]
Thanks @polar spade
@quiet solar Thanks for your help too
I think an Interlude is part of a lyrical piece. So "instrumental" (a genre kind of) can be confusing for a song with lyrics, unless used in an extention first.
Also in general try to avoid using too many many meta tags or overly complex ones
Often degrades the sound, sometimes it could even be better to describe the progress in the prompt itself
Hello, New user here. I’m not sure if I am in the right place for my subject concern. But, please kindly point me in the right direction if I am not. My issue is that I have shared a song with family and friends via SUNO and I am able to play that song through the links I’ve shared with them but the song itself is not visible in my library! I’ve submitted a request for help from SUPPORT but one of the avenues they suggested in the automated response was to come here. Can you help me with this?
In Library page - check the filter - maybe it was hidden?
Heading there now thanks!
My filter doesn’t look like the you show here!
Am I in the wrong place for the filter you show?
Yea you are - but in the mobile app
I see All Songs already set - so if it's here it should be shown.
It’s not here but yet I am able to access the song via text messages I’ve previously sent out
It doesn't come up when you searched? Or keep scrolling to find it?
No
At least give it a like. Then filter only liked songs. See if it shows.
I typed in the actual title as it appears in to the search and no luck.
I try the “like” method
Nope that didn’t work either! It was already marked “liked”😣
I'm having problems with Suno's cover/extend feature on an audio I want to add additional lyrics to it while wanting to keep it's original instrumental and vocals but with additional lyrics, but it actually gives me COMPLETELY DIFFERENT vocals and instrumentals than what I want
Gotta ask - is it your song? or shared by someone else in the family 🙂
@orchid radish yes, ask the how-to questions here
Well, normally we can extend, put in new lyrics, paste in the same prompts from the original output we're extending onto and it would work well enough.
@half dragon no it’s ok… it is my song.
Ok, I understand, but I used both extending and covering of an audio that has vocals and instrumental In it, added additional lyrics, and even tweaked the audio influence settings, but the vocals and instrumentals change from the original, an audio I used has calm, soft vocals, but the extended/covers of it gives me rock-like strong, powerful vocals, and the instrumental changes too
Usually Extend would keep it consistent better than Covers.
Covers for the most part would maintain the melody but different performance.
I also tried to publish the song via the link and all it does is give me the processing “spin”
Try the other way - dislike it and set the filter and see if it shows up.
Wait - no dislike filter on app
nvm
Have you updated and restarted the app? Some of my songs disappeared until I updated and opened the app again. As soon as that was done they reappeared.
I would also check on a computer to see if the song is in your library that way. If it is, then it’s just an app issue.
@lavish coral ok thanks! I will try that right away. @half dragon @polar spade @lavish coral thank you all for jumping right in to help. I was as nervous a long tailed cat in a room full of rocking chairs😊
With you all I’m not as much anymore. Thanks
@lavish coral Does logging off and logging back in initiate the update because I don’t see a specific indicator for an update anywhere. Where should I look to initiate the Suno app update?
Are you using an iPhone? There was an update in the App Store about 4-5 days ago
@lavish coral Yes, I do have an iPhone. Cool, I will go and check the App Store. Thanks again!
Do you expose/support API usage? I don't see any API docs.
@quaint moss ?
No. There is no such thing.
@lavish coral just opened the SUNO app on my laptop and same thing no song. It’s just strange I can play the song via texted messages I’ve sent. Hopefully SUNO support will have answers if I don’t find a solution here.
It's also possible that it's in your trash bin. I know that the link will still work.
@lavish coral @half dragon @polar spade
I’ve found it!!!!!!
So the desktop version obviously has more options than my mobile app. When I went to history, it was there. And it allowed me to put it back into my library.
So it was indeed in the trash bin or Archive as they call it now aday
@half dragon
Morning/evening/noon. Anyone used the Editor or Studio recently? I couldn't make either of them work. The workspace that shows track(s) turns into a white rectangle almost immediately. Is it local to me, or borked globally?
Hi could you help if alls good?
...if you read Dasvon's last message, I think he already told you pretty much the core of how to do it. xD
The V5 Remaster is a joke. Now I took a V5 version of my own songs and with the high remaster version it totally changed chords and melody… A remaster shall give a better sounding version, not a complete different song…
Go for Low/Normal instead of High, because the higher the worse youll get😆
getting a 404 error messages in editing. doesnt save as new song
I almost always put [Guitar Interlude] or [Guitar Break] between the 1st chorus and 2nd verse to give the singer time to breathe.
[Long Instrumental Interlude 1]
[A 20-second instrumental section where the drums frantic] took a bit to figure out. add the instrument, I love drum solos, what fits your sound
How do i get the vocal stem to be more clearer?
Any key words to get consecutive blast beats?
if you were gonna describe backrooms songs like metahesh - i might be dead, whats the prompt
I keep trying to get it to do spoken word and it keeps failing
[spoken word]
"things you want spoken"
do that in your lyric block
like this: [Bridge]
[Spoken word]
"I cast Fireball!"
“You’re in a library.”
“…I cast Fireball.”
“Okay, well now the town’s on fire and you’re all level one—good job.”
will try that its been a weird thing some come out right this one is having such an issue
it works pretty consistently for me. do you have anything in your excluded styles?
so much I tried to just go crazy and repeate myself it didnt work
https://suno.com/s/A7qjAATcJIqsVL2q 1:50 in the song
I have to remeber doing these this worked!
then it didnt haha will have to edit it
I had to do it for each section and then it started to fail and sing again
haha oh ai
thanks tho @green cairn
welcome! and good luck!
thank you fellow RPG player!
anyone know of a concrete way to make songs not cut off super early? lofi specifically seems to want to be very short
instrumental lofi in particular
for yours i tend to have the oposite issue i want a way to create short sounds at times
would be good to have a "target length" tag
right now I am trying to edit sections and it keeps playing random songs in the backgroud
there must be a bug
I second this
Yeah, that’s for sure 😅.
I need help if anyone knows how we can reliably get a delay lyrics start:
First line, (chorus) needs to wait a brief moment to start
example: https://suno.com/song/86ef5e81-3e02-48b0-ba86-b83e58992ac0
what ends up happening more often than not: https://suno.com/song/1ebadb9b-bdd6-437e-a26e-ece84cc4a237
Which is freaking trash. I would like to know if anyone knows how to reliably delay the start of the lyrics (fyi, I'm working on a .wav of self produced instrumentals so the covered instrumentals is the same every time.)
Help. I am trying to do a ‘narration only’ piece and no matter what I do suno insists on adding someone humming or something else. So frustrating.
This is the first step of a multi step process. And I only want a male narrator but suno insists on adding a female voice.
Adding [Build] before the [Chorus] to build up a bit before starting the Chorus might have more chance?
Or try [Intro] ??
10 generations, 4 success with the [Build] tag before [chorus]. significant improvement from my original 2 out of 20 I can work with these odds I appreciate the assist fam
style -deep full male choir,,,,, What i got Humming Mini Mouse ..... Its like throwing credits out the window- V5 is supposed to be better ... better luck with slot machines it seems
why does suno suck so hard at doing unplugged acoustic guitar? I've tried like every variation of "unplugged single acoustic fingerstyle guitar" "simple lullaby" "bedroom recording" "analog" etc etc, and it gives me a symphony of instruments and polished electric-acoustic generic elevator music.
Suno v4.5 played very nicely with audio recordings of acoustic guitars - replicated the strumming perfectly, and even picked up on hammer-ons and really subtle chord modifications. So that may be an option if you have a guitar yourself and can strum the basics of what you want.
Alternatively, you could try Suno's direct interpretation of one of my acoustic guitar recordings as a possible starting point for a prompt. I've not tried prompting for this, but in principle it contains the right kind of steer:
"Acoustic [your genre here] song with a male vocalist. The song is in a minor key with a moderate tempo around 120 BPM, The primary instruments are an acoustic guitar playing a fingerpicked arpeggiated pattern, and a male vocalist with a clear, slightly melancholic tone. There are no complex production elements, relying on a clean, natural sound, The acoustic guitar provides a steady rhythmic and harmonic foundation; the vocals are delivered with a narrative quality, clear and forward in the mix."
Well basically i just want something that sounds like it was played on a single unplugged acoustic guitar. Fingerpicking folk style. Recorded on the phone in a bedroom or whatever. Lo-fi. SUNO doesn't seem to understand the concept. It will make it sound like the guitar is plugged into an amp and it will add a drum machine or something else. It's biased toward making everything sound produced and professional and to add effects, like echo, and to add multiple instruments instead of the single one in the prompt. What I also notice, when using the instrumentals prompt in SIMPLE, is that SUNO will just go and write its own STYLES prompt for it, which often disregards the prompt I wrote. Like the word 'fingerpicking' or 'fingerstyle' for example. So I use STYLES instead and that doesn't understand solo instrumental at all.
Let's say I created a song and like how it sounds but now I'm missing the melody because it is only containing singing. How can I effectively type in a proper promt so that it will create an instrumental like the singing had.
I tried to use the extent option, with instrumental promt, or no voval promt, but it's still contains singing.
any idea?
I imagine most of Suno's training library is on full production songs, rather than lo-fi recordings, which probably isn't helping. If it thinks there should be drums and professional production effects in the genre, it can be tricky to persuade it that there shouldn't be in your song.
I took a few attempts based on my style prompt alone and just couldn't stop it bringing in a drummer either. So realistically that seems to give you two options:
- Play the basics of the song on acoustic guitar yourself, upload it and ask Suno to cover the song with very high audio influence
- Let it do what it's going to do, then split the stems afterwards (assuming you're on Pro/Premier) and use the Editor to remove every instrument which isn't the acoustic guitar.
Sorry, can I clarify what you're trying to do.
You've got a song which is just a solo singer, and you're trying to create an instrumental version of exactly the same melody with supporting instruments, but no singer at all?
hey guys. idk why but i get always rock vocals
and i wrote a negativ prompt no rock vocals
and i get still rock vocals xD
What's your genre?
deubstep
Are you trying to create a instrumental version of the song, where intsruments play the melody instead of singing?
Okay, that's...weird. Suno generally doesn't like being told what NOT to do, but is quite receptive to being told what to do. Have you tried prompting it for something that would be very clearly dubstep vocals?
i get since 1 hour only rock vocals
i can try to make only the beat
and put after the vocals on it
yes its clearly dubstep
ui i used the auto lycrics no i get normal vocals
n1 tx bro
If you're looking for a prompt, you could try something like "The vocalist has a crisp, rhythmic style with phrases that lock tightly to the beat. The vocalist performs in a dub-step style with occasional ad libs and heavy production effects."
It would take a leap of interpretation for Suno to read that as "give me a rock star".
xD
there is not 1 single word with rock
or something like that
or distooration
or something like that
no metal
no guitars nothing
If you're able to, try doing it in v5 and v4.5. If both of them are struggling, it suggests something is confusing Suno somewhere along the line.
It's not as if dubstep is a completely unheard of musical genre, it shouldn't be struggling that hard to pull in appropriate vocals and instruments.
okay
with the auto lyric mode thing i dont use it very often
i get normal vocals
"normal"
Auto lyrics shouldn't make any difference - that's just how the lyrics are created if you don't have anything to put in yourself. What affects the music and singing is in the style box, or any style prompts you've added into the lyrics box to give specific direction for certain parts of the song.
why 5v remasters have quality poorer than the original soong
That's a potential headache with remasters across all versions. Generally, remastering works best if you started with something in, say, v5 and remaster it using v5. Essentially "keep this mostly the same, but have another go at it". Sometimes it'll be an improvement, sometimes not. It tends to be more hit and miss when the remaster version is different to the original version.
Unfortunately, Remaster isn't a magic "make my song better" button. At least, not consistently.
yes
i tried the remaster function and most of the time i dont remasterxD
im remastering 5v songs
it just turn songs worse
sometimes he spit out something good
on my last track he makes his job good
but on other tracks the complelty destroy it xD
now is extensions that arent working lmao
it cant spell anything
just horror beyond human comprehension
why isnt working
it was fine like
a week ago
why not now
suno is the first AI that gets worse
No, it was a regular song but turned out without melody in between playing
nah
I am able to edit the song later on since I use professional tools, but now I don't have a matching melody (instrumental part) which I can add in front of the song or in the middle.
I'll be honest, I'm still not 100% sure what the issue is. Would you be able to DM me a link to the song? I think hearing it would make this a lot easier to understand.
Maybe not mentioned yet, but you cant use 'negative" prompts. Neither " - " or " no"
We are not quite prompting for "guitar" when we prompt for a guitar style, as Suno looks around the target of "guitar" and give the most optimal sound that would work for whats being asked with the rest of the styles and lyrics, if any. In that case, likely a guitar.
So you may not get a Stratocaster guitar by prompting "Stratocaster guitar" unless the rest of the piece had a vibe which might often use a stratocaster guitar. And if you prompt for a "no rock guitar" it doesnt even know the style of a "no rock", so does its best to find the closest rock guitar it can.
Same goes for prompting for a "no rock vocals" . Suno will give you the closest rock vocals it can fit to work with your music piece.
You may need to try different combos of original production and remastering as well.
Prompt in 4.5, ,then remaster in 4.5 , then remaster "subtle" in 5.0
Or remaster a 5.0 original with 4.5 first then try any setting to remaster into 5.0.
Does one of you Prompt-Wizards have some tips for what I should put in the Lyrics Box of an Instrumental Cover?
Wouldn't really make all that much sense for me to write some guiding Metatags in there but maybe someone here knows of some good ones that could work for controlling the soundscape a bit better
Since Im no prompt wizard, Id say go wild.
But tips hmmmmm:
One is.....Keep it few. No more than 5. See how prompts control the vibe by moving one to the left, or to the end, or front.
Style prompts set up whats to come and , kind of, where when we indicate a similar prompts in a metatag. So having "guitar-prominent" in the style box set up Suno looking for a metatag "guitar solo" in the lyrics to show Suno where to put a guitar. This is not perfect, by the way, as Suno then may stick another guitar riff after the second chorus where ytou didnt want. But thats what editors are for.
OK. Theres two. 🙂
oh this is quite insightful already!
As for control sound with metatags, that might be more of a Google search or search in here for for posts over the past years. Thats the better best to then experiment with those "fidelity" style prompting to do , such as , "clear vocals " . There are likely hundreds by now.
I have a wiki page on hand that is quite big only dedicated to this. I am quite overwhelmed I must say lol
Yes. Thats my point. Lol. Too much that others think will work, and maybe they have some influence.
Thing is, your genre (or mixed genre) can affect things, your lyrics can insist on a scream rather than a whisper, you metatags or style prompts might be ignored if the genre is just too far from applying what you're asking for.
And dont forget to just try the remastering tools, or cover what you have first-- this helps test what Suno is hearing in your present masterpiece -- see how its working and maybe needing to trash most of it , starting over with just the first minute and extend several times with style specific prompts each extension in order to keep the fidelity in the direction you want.
But as for specific prompts, im not the guy for that, I guess, when given a genre, which you havent mentioned yet. 🤔
It's a Video Game Theme, though I find that hard to prompt, it's overarching Genre would probably Hard Rock or Metal instrumental with some Brass or Synth work in the background.
Rather than just post the link, here is the text from @hollow hare (from Oct 14th) that is a good background to learn from. I typed "muffled" in the discord search of this channel and saw this:
" . . . When trying to use terms for exclusion, you have to think of what terms it has been trained on. Just because AI can predict what it thinks you want to hear does not mean that itself knows ahat you really mean.
E.g. Once you realize that the odds of having the term "muffled" or "cut off" part of the training is infinitesimal, you would not be surprised that it did not work.
You have to rethink on how to change your prompt to get what you want. Some tips I usually give:
- Be specific in the subgenre of what you want. The more specific you are, the smaller the learning pool which leads to better quality (especially if you have a genre with diversive styles, like with pop)
- Adding in emotion helps get what you want. E.g. melancholic, happy, joyful
- If you want longer songs, add in interludes or instrumental breaks, like how most radio songs have. E.g. [Catchy hook], [Instrumental break], [guitar interlude], [Melodic buildup], etc.
- You can try adding in qualitative keywords to help get more clearer learning patterns, like "high definition", "studio", etc. Be careful of some terms, like "vinyl" may add in that vinyl scratch sound, "cassette" may give worse (or traditional) cassette quality sound, etc.
What the exclude prompt useful is for things that have good odds of appearing, for example:
- Even if you define male vocals, there is a chance of female vocals or duets
- Adding female vocals to exclude will help further reduce the chance of that happening . . ."
Neat
I searched "video games" and a mention of using chatgpt to give suggestions of prompts for a popular game designer. #📚┃suno-school message
Also, now knowing your genre, go to the Suno Home page and search for game music, listen to a few for some favorites to see how they are prompting.
That might be a gold tip
Thanks a ton for these
Give it about 6 hours and post your question again. That'll give time for others to wake up and on a lunch break in the USA and for Europe to be eating dinner. Good time to get more responses.
Hope this little bit helps.
I was busy trying to implement your mass of wisdom into my prompting, I didn't mean to sound rude or dismissive like that. I really appreciate all the time you took to write this to me.
Didnt seem rude to me. I didnt notice. So its all good. No worries then. Good luck.
sure once im on pc i ll send u
I'm uploading a song I made while singing and playing piano. Then I want Suno to add music. I want a neo-soul-like groove and it hits perfectly in the first verse and chorus, but then it starts to overplay with drums doing 1000 fills, singers doing adlibs and brass going crazy. How can I get Suno to stop overplaying using V5?
This might mentioned before, but has anyone experiencing low audio quality and noise like hissing? My experience is in deep house music and everyday is getting worse instead of getting better.
Hi, I need help. I renewed my Pro subscription in the Suno app for Android using Google Pay for $10. The $10 was charged from my card, and I received the credits, but I did not receive the subscription itself. I can’t use V5. When I try to resubscribe, it shows that I am already subscribed, but the system still asks me to upgrade. I contacted support, but they haven’t responded.
make extensions work again
it cant spell any word
its unusable
i lost more than 1000 credits already
Check your lyrics for errors in the meta-tags or parentheses.
Try different weirdness/style/audio slider levels.
Try your extension at different starting points.
Check if the original song you want to extend from already has errors in the lyrics/pronunciations.
Check the lyrics alignment in your original song in editor.
Use additional style tags for your the singers language/pronunciation etc.
Try&Error, with different options.
You can also try model v4.5. Suno v5 is still beta and is known to have problems with "mumbling".
man i tried 1000 credits
i tried EVERYTHING
its literally just doesnt work
its even more bugged now
it cant even song the entire section i wrote
just spell horrors beyond human comprehension and hope me that you will be happy with this
everything from my recommendations, too?
We can only guess if you do not show any prompts 🤷♂️
yes
i used suno as a professional for 10 months from now
i know what to do
and i know its broken
Ok, why ranting here then and not submit a bug ticket?
i dont know how to do this
i always complain here and someone appears to solve it
If only everything could be so easy! 😉
but i lost already 50% of this month credits
and my channel needs content
without full songs i cant finish anything
past month i did over 13 full songs
this month is so bugged i didnt finish the first one
i dont know how they messed suno but i will hope they solve it soon
It looks like you already posted in https://discord.com/channels/1069381916492562582/1082775596767117312
So, you have to wait.
So far I did not have any unsolvable problems with extensions.
You could also try an instrumental, extend it to the length you need (but below 5min). And then cover it with lyrics....
my song is narrative
i dont do this kind of stuff
it doesn't matter if you cover the instrumental song with lyrics in a second step.
the first part is concluded
i cant mess with it
i just need the second part
but suno hates me
why not? your song, your decision. you could even cover/remaster it and then extend it again.
i sent an email
any one having issues with track glitching out of sync like it went back 1 bar and starts agin?
bit buggy this V5..
Yes, can happen
Its doing it a lot.. wasting credits...
Contact support@suno.com if you haven't - the only way to resolve account issue.
I notice I got skipping issue a lot when using Inspiration mode - not sure if it's the same thing as you experienced.
Reported every clips so hopefully it'd be fixed.
does anyone know why when i open suno editor it starts playing a song, (not the song I opened) and i can't stop it?
how do i keep the style the same when i change a couple words in the lyrics in studio?
Sounds like a glitch or bug - which Editor? Legacy Editor?
Some has experienced it and if I recalled refreshing would stop it #📚┃suno-school message
not legacy editor
Yep its happening to me and other ppl also. For few days already. Not fixed yet😬
Nah, when refreshing it plays other songs😅
Are you using mobile app? iOS? as I opened a few songs on Editor just now and it didn't happen to me on PC browser.
Open in Editor isn't available in my android app.
Browser, tried Opera, Chrome, tried to log out and log in and still same issue🤔
On pc
Interesting. We must have different versions.
Have you check that "play like songs" is turned off. Just to stop the nuisance until something is fixed, that is, if anything. And yes, i think it may even default back occassionally that we have to turn off again.
i dont know whats happening to suno
to much bugs since 5v launched
it was perfect on 4.5+ era
but even 4.5+ is messed now
where is "play like songs"?
See the photo. Click the hamburger menu (1) on a playing song banner, then click the infinite loop (2) on the popup Playlist.
Edit..." play related songs"
As always for many sites and DAW, if on desktop, tapping space bar will on/off ▶️ , if clicking the ▶️ gets buggy.
thanks. next question, why when i do cover and change the lyrics does it keep saying a few words from the original song and not my NEW lyrics?
Oh thanks, when i turned Loop it didnt fixed, so i tried to turn off both buttons, and its working fine now🥳
Thats an old glitch, it may have raised its ugly head again. So the final output has lyrics truncated, but the output is playing OK? You can hear the full lyrics, I think?
Thats definitely a bug/glitch. My workaround then was to diligently remember to copy my lyrics, often using Legacy, so I could copy the edited lyrics before clicking Execute and past edited lyrics into outputs when done.
Then, if I recall, bring back into Legacy and check alignment again. Its been awhile, but yes , sorry to say, I had to workaround it a while until it was fixed.
It may be a small side effect of something else they are updating. 🤷
Because the original song is being used by Suno as an audio prompt. Sometimes that bleeds through and Suno accidentally prioritises the original sound rather than the edited lyrics, or fuses them together. More recent versions have got better at avoiding this, but it happens from time to time.
@glad ibex Yes. It will sometimes stay true to the term "cover" and well, "clone" a little back into the mix. 😄 I like it sometimes, depending on what im creating.
And when it clones my voice or guitar 🎸 sometimes, its fun to work with latwr on making personas 🤫 . So I dont complain. 😊
thank you, finally i'm getting some answers 🙂
@quiet solar burning the midnight oil to give us the latest. I just read their comments to learn all this.🙌
can anyone help me? I cover the song to get a good quality instrumental but it cut off a verse. i spent a lot of credits but the result still the same. is there a solve for this problem?
https://discord.com/channels/1069381916492562582/1372994765196361774 has a few good reads to know while working in Suno.
One being this one ----> discussing short cover output
https://discord.com/channels/1069381916492562582/1422252931943501935
thanks
Suno does better to cover a shorter less than 4 minute song, anywhere from 3 to 7 minutes. Any longer songs than 5 minutes may get truncated to 3 or 4 minutes. Sometimes, Cover will Cover a 1 minute song into over 7 minutes, so its depends on how much it needs to allocate for seeding the new output.
then, is there anyway to improve stem quality? thought I put the studio prompt when generate, but it still unsmooth without vocal.
it would be great if you could put accent marks or phonetics to let it know exactly how you want the word pronounced
you can
I did not quite follow the question. Sorry.
What does unsmooth mean in this case. Where are you putting prompts to then generate a stem? They seem to be related in your question. Can you explain another way?
And believe me, im old, so I may simply not follow at times what someone means.
Also, im at work, so any response may be delayed a while.
my bad. i'm not very fluent in English. I mean when I separate stems to get the instrumental. the sound is not smooth. it's like extracting vocal from vocalremover ai. the sound feel hollow, sometime loud, sometime low.
That's because they're not really stems in the traditional sense.
Suno creates everything all at once when it generates the song. Stems are then retrospectively split apart similarly to a vocal remover afterwards. Which is why you can hear those exact same types of issues.
thank you. is there the way to fix it? I want a good quality instrumental also.
suno isnt working anymore
i give up
gonna find another Ai
5v is the worst thing ever happened to suno
You can try extracting vocals using your own preferred vocal remover, which may or may not give better results.
Other than that, there is no fix because there is no original 'clean' instrumental stem to go back to.
I see. thank you so much
Did someone ever managed to give a voice a british accent. Regardless what I write or do the voice sounds always American.
Yes I have tried this a while back on version 4/4.5 I put the instruction in as part of my prompt in the lyric sheet as well as the style prompt box.
[Vocal style: female UK vocal style, heavy or subtle British accent etc] Sometimes it works sometimes not. You probably need to make multiple generations
I have used a chat GTP lyric writer as a guide to writing prompts for this kind of specific instruction
I guess I need to try further.
Using "Remaster" on a stem can sometimes help with this (the default "normal" strength is good to try first.)
You can also stem the stem ( use the vocal/instrumental for less credits, though Ill use the multi-stemmer which sometimes separates noise into FX FILES) , and see if it filters more to make it better, or it makes it worse. Then even Remaster ( once or again) as @reef pendant mentions.
Wanted to ask, is there a better way to make suno sing a line in a male / female voice. ive been trying F: and M: to say which line should be which with duet in the prompt.
I got Britishy-ish , but it could be New Foundland dialect (with the music having an Australian vibe) , just using femme folk , plastic punk. https://suno.com/s/zNCtv9xy4SGsi1B7
But ive noticed punk, especially "80s punk " get British dialects often.
Short answer: Unfortunately, no. Not reliably anyway.
There are plenty of discussions on this and options to try. One being to use Extend to build up the duet and consider using different Personas. But no matter what technique used, making duet is going to take a lot of trials and errors and lots of credits and patience.
Can anyone tell me a reliable way to make instrumentals longer than 3 minutes? I've been told things like "add non verbal character to lyrics" or "add a length requirement in the prompt" but nothing works. Every song I make defaults to about 3 minutes.
What if you create a really long structure in the lyric box? something like multiple sections of [Instrumental] between [Intro] and [End]? Not sure how long it'd need to be more than 3 minutes but can always experiment. I know we can't specifically set the length via prompt.
I get longer instrumentals by using symbols and dots in the lyrics window. Most important is the genre you are using, try not to use too many different styles that Suno needs to blend, and describe the type of instrumental in the prompt window, along with any breaks/solos/ect. Then do not select the instrumental switch and enter a number of lines of symbols or matrix dots in the same fashion you would written lyrics. This give Suno something to work with, and the more you give it, the longer a track tends to be. I have examples of this I could point you towards.
I added " LP album" and "deep cut" to a medium song and got over 4 and 5 minute outputs on 4.5 . May be a fluke, but there you go.
But even a 3.5 min song, added a 2.2 minute extension. About 6. mins total.. Works for me. And even sounds like the Cranberries. Cool. 😎
i'm working on a song in hebrew and spending A LOT of time trying to get each word pronounced right. I find it's easier just to generate a new cover than fix a few words. If anyone knows how to fix individual words in editor or studio please tell me. for example, i separate the tracks into stems and want to change just the vocals so the singer says "hee" and not "heh". it shouldn't be that hard!
Phonetic spelling usually works.
There are a number of word even in English that get mispronounced, and they almost always are corrected with phonetic spelling.
what do you mean by phonetic spelling? spelling it like it sounds or using special phonetic spelling symbols?
i got it almost perfect by changing the way i spell the words, i spell the hebrew words in english (phonetically), i just have a couple words i want to change. how do i do it in studio?
why can't i see my lyrics in studio and what is that box in the middle of the screen that you can select the instruments and it says style and lyrics? i want to change the lyrics and keep everything the same.
Spelling the word like it sounds... an example would be... Grocery.. phonetic would be Grow-cery.
I would extract the Stems before going into Studio, add all the stems, and on the vocal stem, select the section you want to change the word - then in the box , select Vocal, put in a style if you want or leave it empty- then put in new lyric in the lyric box then click Replace
I only have a pro account so I have no experience with Studio at this point.
why can't i see my lyrics in studio
You can - Press number 2 key for shortcut to show the LIbrary view, select a song and if the details panel isn't shown, press number 4, the song lyric will be displayed there.
If you select the clip of the song on the timeline and if you don't see the lyric, on the top of the info panel, there are 2 buttons, Clip and Track, select Clip and you will see the lyric
why no song has 8 minutes anymoew
I got several songs in 7+ minutes - I think if you have more lyrics, it'd go longer
my lyrics are worth 13 minutes
song stops exactly at 7:27
Did you also use reference Audiio or Cover from another song? That might be the factor too...
no
suno is bugged and limited
thats why
You have enough for 13 minutes, might be a clue. Suno still has to decide an optimum tempo and the actually end the song to determine the overall length of the output. Even if we rationalize "hey I didnt exceed the lyric charact limit", we still infer a genre or style which delegates some say in the tempo speed.
Its not geared to just keep singing until the 8 minute mark cuts it off.
im not dumb
im putting enough for the 8 minutes they promise me
and the remaining i will use extension
The max caaap for aa single gen is 8 minutes ( i believe) so even if it's 13 minutes worth of lyrics (not sure how one would measure that), then 7:27 is doing a good job of trying to make a song. if you want it longer, need to break your song structure down into sections and try creating with multipe extends, using lots of [instrumental breaks] to create more music in between the lyrics in your song structure.
The promise of "Up to 8 minutes" means anywhere from 1 second to 8 minutes. But truly, songs seem to be between 30 seconds and 7:59
2 weeks ago i didnt had this problem
suno is broken
for 2 whole weeks
and no one solve it
Yes. We can extend past 8 minutes. Almost without limit, but more thana 12 minutes seems to put a strain.
no
extensions arent working either
nothing working
suno is dead
Ahhh.... then I'm mistaken. My apologies.
Hope they work out the bugs and get things back as you had them before.
i need to post on my channel this week
but im afraid its impossible
suno wont colaborate
I just did a 5 minute song and a 3 minutes extension an hour ago. So it could be relative to many factors. But a bug or glitch could be part of it as well.
Suno growing fast, I support them and just work around any issues.
The use of Persona can also reduce the maximum length (as appears to be the case in this screenshot.)
suno changed it?
i always use personas and the lenght stays fixed at 8 min
This has been the case since Persona and Cover were introduced in v3.5
i have already more than 15 songs published with more than 12 min
always used persona
and worked and not ripped off my lyrics or did horror beyond human compreension melodics
like its doing rn
is suno overwhelmed?
How is the difference between 7:28 and 8 minutes songs affecting how the extensions work? Do you feel they are related.
Personally , it dont find it much of impact whether my song goes 7.5 minutes or 8, especially if im extending further. So im looking for the rationale of not working around it until it changes back to the status quo before.
extensions arent work from months rn
it doesnt even is respecting the sliders
So its the extensions that are more of an issue?
extensions broken
sliders broken
generation broken
replaces broken
What does "not working" mean? Your getting 2 second extensions? Thats a common one. Or something else?
i dont thing anything is working fine
horror beyod human compreension lyrics
ignore my lyrics and sing what we will hear in the hell
i think personas are working fine
but only that
when it gets acceptable again i will not have any credits
and i dont think i will ever have these bugged generations credits again
If it were "free" free then the issues would be better tolerated and less broken.
Sorry I cant help there. I remember the feeling last year of running out of credits.
Alright. I won't say more then. Looks like a quandary for sure. Extensions work on my end, so some corners may need a bit of cutting if we want to find any workaround that gets momentum.
But I understand these are your creations, even when half of Suno is in fluctuation and beta. Ill just throw in 2 cents worth if I feel I can help next time.
Good luck.
i will block myself of accesing the website for at least 3 days
im afraid that my listeners will hate me for 2 weeks without songs
but theres nothing i can do if suno hates me
im not a talented musician to do it myself
Honestly, I dont like seeing anyone stressed. So might be something healthy to do as well. Your listeners should still stick around when you get back. It will entice them to eagerly wait for whats next!
these song im making is very special for me
cause its the AI character that represents my channel
not literally an ai character
One thing that you might try is a flush of all your browser cache and cookies. I had really odd issues with things in Suno not working or working partially, or even opposite of the way it was supposed to function. Turned out there were a number of corrupted cookies that Suno relys on that caused all my issues.
a real game character thats an AI
Yes. I sense that. Its more than I could do. So keep going!
Some advise I can lend, dont put all of your production into a single tool, have backups becuase nothing in software or comute is completely safe from crashes and corrupted data. Always have a backup plan for when one tool stops working.
Recovering deleted message (let me know if you want this removed):
Message sent by @polar spade Deleted in #📚┃suno-school
Its not easy to chat about "broken" . Any more details than that?
But if you wish to simply have it work as before. I get that frustration as well. The history of Suno is that sometimes it gets funky before getting better. But still , using workarounds seem to move everyone forward.
Hahaha. I added a question mark and .....poof! But an old message. Thanks.
Hi guys, i am new to suno, but i have a question.. when i do a longer song for example 3-4 minutes, i cannot get like a only one solo artist, each time at some point i get some kind on chorus or backing vocals, is there a way I can wrapp a condition to the lyrics, i see that there is a tag [verse 1] and so on but not sure how this works.... Thank you
i tried another browser
nothing changes
the AI is broken
On the same machine? I would still suggest clearing all of your caches and cookies for the entire machine.
i tried the android app too
same erros
erros
errors
the ai itself is broken
man my subscribers will kill me i promised 9 songs this month
Ah, ok. it could also be a problem with your account, I have seen other users have odd behaviour issues with Suno and it took Suno to fix their account.
Unfortunate.. But again, If you are doing somethign professionally, you really should have backup plans for when tools dont work.
Does it need to be v5? would v4.5 unblock you? Does v.4.5 still work the way you wanted? Maybe?
Honestly, I still prefer 4.5+ for some of the genre I produce, and even v4 for a couple.
Good evening, I'm glad to be here with you.
It's weird that Suno would be so totally and specifically broken just for you.
Is there anything that you're consistently doing differently to someone dropping in and putting in a prompt and pressing "Create"? You're not using Personas from pre-4.5, or anything else that's known to cause widespread problems?
It happens, we saw that happening with specific users some months ago, turned out some of them had account issues. Took Suno CS to fix it.
I also had problems when v5 came out, but it all depends on the tags we ask the application to use.
for example I use tags like this:country blues, deep male vocal, gravelly baritone, emotional slow blues, soulful delivery, whiskey-warm tone, raw masculine voice, heartfelt storytelling, melancholic mood, intimate atmosphere, acoustic
with these tags I create country music
4.5+ and v5 both benefit now from more defined prompting, simple genre and styles generally dont work the way they did in v4 or earlier.
I agree with you, but try telling him what you want through tags...you'll be surprised
2 years here, seen all the meta tag evolution 😄
😅⚘️ I don't have that much....but I have over 1.5k songs created from v3 to v5
Thank you
Thank you too
Presently cant download from one of those "other" AIs. So just be cautious if you do.
What's your point?
hey question.
i'm sure some of you know who t-grisly is.
basically, i'd like to get like the synth bass line he uses in his music. what stiles do i put in the stiles prompt field?
YOu could try...
Psychedelic hip-hop beat with thick rubbery detuned analog Moog bass, neo-psychedelia, wonky synth-funk bass, sidechained bass.
And it's T-Grizzly 😉
aw and i put that into the simple prompt thingy, right?
it's Tee Grizzley 😉
How to stop SUNO add Vocalise on start 🫠 🫠 🫠 🫠
Putting an [Intro - Instrumental] tag at the top might help.
Hi everyone! I had a question: why when I try to extend a song on both versions 5 and 4.5, it only extends for 2 seconds, or 10 seconds at most? Maybe I'm doing something wrong, but I remember it working fine a while ago
Try starting your extend from the middle of teh previous stanza, if you are just trying to extend from the end of the track it usually just ends it again in silence or a couple notes.
Enter the last of the verse or chorus you are breaking into and then add your additional tags and lyrics.
almost always even from the middle of the song or from other parts of the song it gives me a max 10-20 seconds
How are you extending the song, and with what? Is it just an instrumental or does the song have lyrics? What style/genre is it?
would offer the 2 cents of do not use any form of [Outro] tagging, or anything that signifies the song should be coming to an end, until you're extending the ending.
For now, I'm focusing on some important musical tracks as covers. I also tried with some of my own original tracks, even with lyrics, but I didn't put much effort into it, but the result was the same. As for the genre, I don't know; there are so many different ones, it depends on the song: symphonic metal, jazz, funk, etc
I never use it, I even tried with some prompts with a longer duration than the song but the duration is always around 3:30
"sosme prompts with a longer duration than the song", do you have something in your prompt that is instructing it how long song should be? I don't know what you mean by "longer duration than song"
No, I didn't mean that, maybe I explained myself badly, I don't speak English well XD
ok, i was gonna suggest taking that part out because it could be having an adverse affect on you trying to extend.
Some people seem to be having trouble, more than usual, with extends. If you know the basic troubleshooting and workarounds try those and hopefully get some better gens.
I would try moving the track you working on to a New Workplace. I can't prove it, but it seems the workplace affects the generations within it. Can't hurt to try.
what do you mean by moving the track working on the a new workplace?
its only on PC/desktop browser. if you're on mobile i don't think you can manage workplaces
yes i work only pc
but i dont udersantd :/
Move the song to a new workplace, when you click the 3 dots select Move to Workplace, then create a new one. This way it is only your track you are trying to make longer in the workplace.
hopefully by doing this all prompt changes you make in an attempt to be longer generations will be understood. Maybe Suno is looking at all the songs in your workplace to decide how to create. So it being by itself gives you a blank slate (in theory).
Wwworkspace* My English isn't great either, and it's my main language. Well, typos might be my main language. Sorry if didn't understand. Just trying to help.
Im not following the thread exactly, but given all things, I just pick from one of the 2 to 10 second extensions, provided it has any reasonable audio in it, and extend off of that. It tend to break the cycle and give at least a 1.5 minute audio so that I can move forward.
I dont bother anymore after two generations of similar short extensions. For a total of 20 to 30 credits, extend from one and keep going.
if i write no wet effects or reverb or delay does the prompt is to strong that he change the sounds?
or white noise
or any noise
Don't use "no" in the prompt - it can often backfire. Try something like dry mix instead if you want to try and filter out reverb/echo/delays
where i write dry mix in the style?
yeah
Anything in the styles window is treated as somethign you want. If dont want somethign it needs to go in teh advanced window in the Exclude section.
i mean i dont have anyproblem with "room" or space or any wet things but sometimes its tooo much
yeah but in the style section its a positiv prompt? i can only push things?
yeah nvm i can try
Yez, styles is only what you want.
that's the best way to look at it, yeah. You're basically saying "I want these things", which is why using "no" is generally a bad idea
Keep in mind that negative prompts can and are often ignored.
okay
yeah for example if i dont want a rock vocal
i write it on the negativ prompt
okay tx for help i try it
ah yes
i would need more space in the style 1000 words are too lessxD
maybe 2k words are fine
Use the macic wand
'Do not never say no' . Ironic , but actually quite accurate. 😊
who?
?
sry i woke up for half hour my brain dont works
still yet need more coffein
coffee
magic wand?
Yes. That wand, will create creative descriptive from your prompts.
fun fact: I used to think that icon was an RPG dagger
xD
like in prince of persia
i think he only optimize the prompt but i think he also delete maybe impoirtant things
idk
okay i try the magic thing once
ahhhhhaaaaaaaaaa
he deleted words
lol yeah it can be a bit finicky. always copy what you have to a notepad or something before making changes
magic always comes with a price
yes xDD
around three minutes long i wrote in the lyric mode this prompt "around thrre minutes long" now all tracks are 3 minutes + -
it was a mistake but looks like it works
for now
Meh. The average song is likely about 3 minutes long. Like saying , make the cows milk white. 😆
you mean farmers don't tell their cows to color the milk chocolate? 🤯 \
xD
how are you this evening @polar spade
My brother was actually quoted in the local newspaper , when he was about 8, that he thought chocolate milk came from the brown cows.
Tha's pretty cool man
lol
what u get on the black cows?
no sir black crows, cows are spotted
cola
lol
^^
i'll go back to wasting credit now
I like convincing kids that mountain goats all have one side of legs as longer than the other. They all think a bit and eventually ask the same question:
What happens if they walk the other way.
devious man just devious
I used to be scared of goats when I was a kid because I thought they were demons. A house in my neighborhood had one in their backyard and I ran home crying when I saw it
Well, technically, they would fall, but that hypothesis requires evaluation and study.
Who is willing to donate to our expedition?
I can see that.
i blame DOOM
wicked game
i go play bf6 now cya have a n1 day
Then you note that if they were on flat land, the goats would all walk in circles. Kids would just split a gut laughing. Yes. Its a night of fun and frivolity at the old household on a Saturday night.
have a good one Paul
Fantastic frivolity on Fridays starts the weekend off on the right foot. (shorter or longer)
or is it the left oe?
Would it be okay if i play the music i make on twitch?
if the track was generated on a paid plan you have commercial rights to the songs, so that would be fine. If you're on a free plan, this is where it gets tricky because a monetized twitch stream may be considered commercial. I'm not 100% sure if that's the case, but it's something to be wary of
I've had a interesting experience with Suno, I wanted to make a cover of a actual real song, but uploading the audio was copyrighted so Suno wouldn't let me.
In turn i was allowed to uload only the instrumental version of the song.
Somehow, despite the origal lyrics never being uploaded because suno wouldn't let me, the AI sung my lyrics in the same way the real version sings it only from having the instrumental version uploaded.
Im more interested in how this is possible and how would the AI know how to sing in the right way to mimic the original song.
you should not be uploading copyrighted material to suno, only original work
Purely gor personal use, i have no intention of using it for anything other then music for me to listen to
personal or not, it's not allowed
I shall get rid if it then
I was jsut wondering how it eas able to know how to sing properly
Given how it only had the instrumental to work off and not any audio of the original lyrics
it's trained on real songs, so if you give it the complete backing track it's obviously going to sing it in the original way because it fits perfectly. just one of the reasons you shouldn't upload/cover copyrighted material
That make sense ye, smart ai.
I got plenty original songs, i just felt like trying one cover tbf
But i know now im not allowed to
I totally understand the appeal of wanting to cover a song by a favourite artist, but yeah, for legal purposes it's not a smart move. keep it original 👍
At least i got to try it once, but ye. One and only that will be, ill stick to original like the rest of my songs.
it's always nice to see how suno can transform an original song. blows my mind every time
it also makes me feel bad because it often sounds better than what I made lmao
Ye. Wild it is
do i can save my setting style lyrics as a preset?
Style? yes.
- To Save it - click on the top of the style box
- To recall - click the icon on the bottom
okay tx
I should click the buttons I never touch....
Except [Instrumental Break] is there others i can try to have a break ?
almosts anything you can think of will work. it interprets langauge.
[Interlude] works nicely
will try thank you
Ive found interlude will get you about 20 seconds. So sometimes I stack.
[interlude]
[interlude]
[interlude]
I have BIN CHECK SYNDROME: Every Tuesday I check my neighbors bins to see if its the right week to put out recyclables. 😅
. . . And good morning all!
.... What did i walk into
AI trains so we'll to both lyrics and notes, that when it breaks them down to random pieces, it still works out that , for instance, the optimal vocal sound that works good with one note, is another note, and often three notes in a row sound good as well, and whats best with three similar notes in a row is one of about 300 phoneme sounds, one might be "jing" , and if jing is on the first note ,optimally, the next vocal sound could be "uhl" . And so on, until it works out, the optimal best vocalization to three same notes in a row may be
"jing uhl bells" or it could be another less optimal combination as its then set to randomize the attempts to give a variety.
It appears, to humans, that its retrieving an actual composition of Jingle Bells, but its has no idea what Jingle Bells is.
Same for chatting with an AI chatbot. It has no idea what its saying and often makes optimal (and sometimes funny or odd) guesses of what the next phoneme in line should be. But it comes out sounding like youre in a conversation.
Thats why Suno still often hallucinates in gibberish. Its giving the best optimal composition and accompanying vocal sounds it can come up with, just as logically we may, not often , but may get "Jingle Bells" on an output.
Well General Chat was deeply entrenched in a convo about Language History, so I didnt wish to barge in with a stupid dad joke. So I posted here instead. 😁
Okay that definitely makes me laugh just now. How are you, Das? Good?
5 x 5 ! 👍
hi, i have a question, is there a tag in v5 that i can place to get only one clean voice accross the entire song? Each time i do a song, if it will exceed 1 minute, a chorus or backing vocals appear that i don't need to have, this can be configured from the lyrics tags, should i switch to a different model, is there a KB regarding tags about this? I cannot rely on pure randomnes on this..
5x5? Really. That's Ham radio or military speech.
5x5 is popular for our older generation for saying
Well it's the same i guess
Like if an older man says that,
It's mean
You and him are you damn similar! 5x5 though

If you search this channel for things like "clear voice" or "clean vocals" you should come up with different posts that could be helpful in some prompts others use.
Suno is all about optimizing randomness. So, though not purely random, we can never rely on perfect consistent results in one or two generations. It might take clicking 30 times to get even one close result that we might like.
Suno even features that Pro users get 10 generations to simultaneously process at the same time as its just innate thst we will likely need that many to compare and get one good output out of twenty choices.
i understand this, but even if i look and search for this, i see people speaking the same, even if they put or exclude backing vocals, they still get backing vocals.. that is why i am asking if there is a knowledge base that you can actually follow and test out the results, with clear instructions on lyrics and on styles, thank you
What's wrong with Cloudflare tonight? It keeps throwing 500 errors.
Oops. I tripped over the plug. Sorry. Will be back up in a moment.
Discord is crowded again
This may take longer than I thought. I cant see in the dark to plug it back in.
👆 Could be intermittently down for the next 3 hours if the maintenance is whats affecting it. Hope not related and up soon.
RIP just when i woke up
that's what you get for waking up
"Get full song" is adding artifacts.
Which suuucks. It means I'll have to manually connect them.
I've never actually used that before. does it merge extensions to the original song?
honestly, yeah.. I don't even bother trying to re-connect to services right after an outage. gonna be rough for a while
Damn, I can't even generate songs now.
Cloudflare still bugs me out
😅
I still have my 286 and modem . Think they have a dial-up number?
I've seen this happen where old uploads (i.e. older than several months) are extended from. Something was fixed with how uploads are handled but old uploads can still have this problem.
If it's not that: Can you send me an example of this please? (The "full song" clip, the extension, and the clip you had extended from.)
My doc's computers are down. Welcome to 2025.
Il sito e off quando tornerà on ?
Dovrai chiedere a Cloudflare. Anch'io sto aspettando e sono ancora più ansioso di te.
Ma cloudflare gli ho scritto senza risposta me in chat ne nelle pagine
Anche perche ha mandato off molti siti vedo
So why not replace CF with Google?
Whom should I turn to for help?
cloudflare崩了
even CloudFare in Europe is down,not just Asia
Really? OMG
Yep im in europe, cant access suno😁
Wasn't with an old uploaded clip. extended from a cover made today. I would link you if the site was up 
I could still get in when my UTC is 21, but not anymore.
/chirp
Chirp is retired, and no longer on discord. You can make music now on our website: https://suno.com
someone like me?
Hi, where should I report Suno app issues
#📱┃bugs-ios #📱┃bugs-android but it might just be a cloudflare problem
oh nvm people are saying the app should work, so yeah, post your bugs in the relevant channel above
Yes it is but my app on iOS never worked properly and bc of the web issue now, I could have used the app
if it's more than a bug and actually blocking you from using Suno, you can submit a ticket by emailing their support: #support-guide message
Thank you 🙏
no prob. good luck
Too many sites relying on one single point of failure. Cloudflare boasts: "Anybody that runs a website should be on the Cloudflare network. Cloudflare speeds up your website by caching content at data centers in 330 cities globally distributed around the world. In addition, our security features protect your website, giving your visitors confidence and may even help search engine rankings. Both of these offerings are entirely free."
But only when Cloudflare works. Maybe they got too popular for their own good.
works for me again now
I would not like to be on the other end of suno support right now
I would not want to be a cloudflare network analyst now..
they're on about $220k. I wouldn't mind it xD
The money isn't enough for the continued stresses of maintenance and crisis intervention. My 35 yrs in IT was plenty
Fair call. yeah I worked as a linux sys admin for a while. Originally I thought it was going to by dream job, but it certainly was not
It seems that cloudflare fixed the problem it works for me now ✌️🙏
It's back!
This prompt is making really cool punchy 2005-8 dad rock sound but it's not fitting my lyrics, if anyone wants that sound
Oops.. I didnt read Woodhouses message fast enough apparently I can t say.... ummm... that itty bitty word. Lol
Seriously, it's fk infuriating; the generated songs keep missing syllables.
Didn't cloudflair and Amazon cloud services go down just last month too? I seem to remember a lot of reports across the net of things not working because of that.
Fck Cloudflare, These damn messes wasted my whole night!
Hello I have a question.
How do I get the AI to listen to my vocal prompts like to do a male and female vocal alternating 😭 I did it before but now it's not working...
It puts female vocals even though I specify [male verse]
Why does it change it to female when there's no prompt until later?
Prompts are suggestions, not commands. Suno can take note of when you specify to change singers, but it won't always do so consistently.
Try using the [Call and Response] tag as an opener in the Lyrics prompt before the section where they alternate, some have found that helpful.
Suggestions not commands... Oof 😭 okay.
I'm still learning to use it. I definitely broke the poor ai brain last month. It started spitting out 15 seconds of really weird noise without using any of the lyrics. I thought I accidentally summoned a demon 😭
Would it be more likely to follow the suggestions if I turn weirdness down or up?
Weirdness down. All the way to 1% .
So, for fun, and it can work sometimes. If you want it to start on a male singing, put "male vocals" as your first prompt in style prompts , and of course metatag [male vocal] before the verse.
Sometimes this will get the alternating patterns in the right order. Even still, every space, every line , every couple of characters also means something and can switch gender tracks.
i want to hope that suno is unusable rn cause 5.5 will release soon
happens 98% of my creations
suno is totally broken
suno looks abandoned to be fair
we need a new model
5v is already unusable
songs
i dont know how suno is doing with souless auto creations, but for custom songs is everything broken
nothing is working
Examples may be useful here?
examples: everything
NOTHING works
suno ignores lyrics
ignore styles
ignore metatagas
ignore personas
extensions doesnt work
replaces doesnt work
nothing is working
i did a test
it cant spell right 15 words without an error
it never happened before
Hard to say, but I'd agree with Dasvon and err on the side of low weirdness.
Generally speaking:
-
Audio influence: the higher it's set, the more Suno will take more direct inspiration from audio uploads (e.g. in Cover, direct upload or persona) and make fewer reinterpretations on the original melody. Setting it high (e.g. 75%+) is useful if you're trying to get it to replicate the previous audio closely, with relatively few melodic changes.
-
Style influence: the higher it's set, the more Suno is likely to take on board your prompts in the Style box. For better or worse. Whether this is a gift or a curse can depend on the level of detail on your prompts. If you're not trying to achieve something very specific with the prompts, mid-range is usually best. Generally turn it down if you're trying to replicate instruments/vocals from an existing prompt.
-
Weirdness: the more weirdness, the more Suno will throw in changes and unexpected aspects across the board. Again, if you're prompting very tightly for something specific, or trying to replicate audio, lower weirdness is helpful. Low-to-mid is a good default, because the model works best when it has some freedom to operate.
3.5 were away superior for this
i dont what happenned to the AI itself
i think people is breaking it so thats always is so horrible
they need to reset the model
Ryan - I asked before, but can you confirm whether you're using Personas and whether the issue still happens if you take the personas off?
happens with and without
and no, im paying for this sheet i WILL use personas
it must working
@halcyon sundial that's kind of what happened to me. I just disliked / reported all of it then deleted them all and then left it alone for a few days. It's working fine now.
im paying for this
In which case, this is realistically something you need to bring up via the support tickets rather than ranting at us in here. We can't fix this for you if it's an engine issue completely unique to you, that isn't affecting anyone else.
nah i just want to spell my rage
Yeah, but this isn't the place for it. Take it to general please.
5k lost credits drive me insane
@halcyon sundial If you're using personas created in v3.5 with higher models, that it tends to break everything all over the place is a known issue, for what it's worth
im using 5v personas with 5v models
it doesnt work cause the ai is pure trash unusable
It was worth asking. I honestly don't know why it's causing you so many problems.
how can i report 500 songs at the same time?
there is another ai?
im considering cancelling suno
I find low weirdness makes generic sounding songs but high weirdness can create nightmares. I usually do about 70% and I've gotten good results. I adjust my lyrics if it's not quite delivering.
The audio influence is only available on the desktop. Not the app. On this note: do people find desktop version to yield better results than the App? I like the app because days when I'm rotting in bed- It's easier to use. It's a little frustrating I can't access the full features on mobile though.
Thanks for the tips! I will try them out. I didn't know there was a discord until I emailed about breaking the AI lol
Hey there! Thanks for reaching out about this!
Would love to better understand this. Feel free to DM me with a brief message about what's going on and I'll do my best to help!
Looking forward to figuring this out!
You just report one by one. I did that. About 100 , probably more. I too used up around 700 credits lol. I should have stopped sooner 😝
ok
I see no issues with this gen?
https://suno.com/s/zuVbL2tfGRNsvpPz
Minimal tweaking just a style prompt and the lyrics.
I suppose I'll add some metatags next to see if that influences thing badly.
I think my issue was I started getting really specific with the prompts which just confused it 🤣 also not understanding how the influence scales worked. Then I got frustrated and made it worse. It happens sometimes 😭
This is why I'll go a while and build up a library of "good prompts" and reuse those, then come back at some point and strip it all down as much as I can to rebuild the structures I use. Over-reliance on magic prompts is a way to get frustrated.
Ah, I tend to use the browser version if I'm working on mobile. I didn't realise it wasn't something you couldn't do in the app.
This one needs some work but there's potential there. Have to try to remove that vocalization at start. Maybe come back in with an outro of some kind?
Now to play with the Personas...
I started doing this too. Adding the prompts that worked well at the bottom of my lyrics.
I don't quite understand personas. I was trying to use Kits AI to train my vocals and then plug in my suno song and it works quite well.
I meant more like maintaining a "cook book" directory...
hey folks, how do u make it so suno only outputs melody, and no drums lol
Did you set your site on a 3'-day hiatus ? You mentioned yesterday.
No no, I understood that. I have been doing that too. I meant in ADDITION to that, I add the prompt that gave me the sound I wanted to the end so I know exactly which one worked in my favor.
Mine is much much smaller and less detailed than yours it seems lol
I think they mean, link an example of a good output and maybe two bad ones with the same recent issues. Or one of your good ones , and then run the same song now to show the issues.
that's a really tricky one, and I'm not even sure how you'd tackle it, because if you use "no drums" or "drum-free" it may detect the word "drum" and give you an 8 minute drum solo. Maybe "beatless"?
I don't know of any reliable way to force Suno to ignore drums, and given that drums are a feature of most songs in most genres it'l probably take some trial and error to avoid them being brought in.
You could try adding variations of drums, percussions, strong beat or rhythm to the 'exclude styles' box, but it's not really designed to filter in or out specific instruments.
It might be easier to approach the problem the other way around. Let Suno create what it wants to, then do a multi-instrument stems split on the song and use Editor or Studio to remove the drum track.
I would like the AI voice to use my voide like swaping it.. dont know how
Suno does not clone a voice really. It would mostly get some vibe, nuance and characteristics of your voice and try for something similar, like a 2nd cousin.
So its just upload and Cover. Audio slider about 85% and the rest leave alone. Often thats a good close character of what you upload. Might need about 6 outputs and pick one or adjust sliders and try again.
There isn't a way to directly do that. Suno will typically replace your voice with an AI singer.
If you have Suno Studio, you can cover the song and try swapping your original singing back in afterwards - try setting audio influence pretty high on the cover, and stating the BPM and key in the style prompt (I'm not 100% sure how much influence that has, but every little helps with something like this to avoid extra editing work afterwards).
Alternatively, as another two-step process, create a cover or new song using your vocals. Then get the instrument only stems of the song and re-record yourself singing over the top of the instruments using whichever external tool(s) you prefer. That has the advantage of not needing to edit your original vocals if the tempo or key are slightly different in Suno's outputs.
As per suno terms.. it states we are just using the platfomr.. tomrrow if any challenge takes place for the osng that i will publish, it will come on my head coz here lyrics are mine but i have use suno musicn and the AI voice.. ant similarity of the voice can be challnged by an artist in near future and then its comes on my head. so to avoid it .. i wanted to to swap the voice with my own.. .. i see a trouble here
And here is the Persona one...
I apologize for the delay, I took the time to grab my lunch, then eat some of it while listening. Screenshot to prove no cherrypicking going on here. This is straight up generating and picking the one I liked the best....
I don't think there's anything wrong with Suno or with v5 right now.
@glad remnant It's fairly unlikely a singer would challenge you for the AI voice sounding too much like their own, unless you're deliberately trying to present your work as if it were theirs.
I am not at all my brother.. i have written a lyrics long long back like 10 years back.. and with suno i have created the music .. whererin little bit in the begining i tried to mix my voice given it suno with the featires.. and then its really suno AI.. i want to publish it.. but then as per suno terms and ocndition that i have read and went through.. it says that tomorrow if smiliarity found and challenged.. its not on suno but on the user as the term clearly states that as a a user i am using the platform..
i can feel it the voice somwhere go close to nickleback... i can share the song here for you guys to give me a feedback.. its one o fmy dream project
here is the song that mix match i have created.. i would really appreaciate a help.. i want o publish it but theni dont wanna come under any legality tomorrow
I don't understand, the songs come out really well, but if I create something similar, battle rap, it always has the same structure: an intro with a voice that does the intro without words, a song structure with reefs that are always with four structurally similar notes. In the first 10-20 songs, everything seems wonderful, but after 50 songs, you start to realize that something is very repetitive. I've tried a thousand settings, even custom ones, but the structure remains the same.
@glad remnant The point of that bit of the TOS is that if you get challenged on copyright, that's on you to defend or respond to accordingly - Suno aren't allowing themselves to get drawn in. That's no different to publishing any music as an independent artist in the first place, however it was created and pulled together.
If you're concerned specifically about the vocals in light of this, then I gave you a couple of suggestions for how you could introduce your own voice in my previous response. Best of luck.
That sounds like the genre itself probably has a lot of repetitive elements in pre-existing songs.
If you write/create your own lyrics, you should be able to specify a different overall song structure through that in terms of Verse/Chorus/Intro/Bridge style elements.
In terms of influencing the melody/notes themselves, there's not an easy way to prompt for that. You can give Suno a chord pattern in the style prompt, and it might take that on board. But it might honestly be easier for you to sing or hum or play on the instrument of your choice the different chord pattern you're after and include it as an audio input when creating the song.
I also tried to help myself with chat gpt to give a different structure, but even if I say that the intro must be "intstrumental" it always starts with wordless choruses and then the song begins, it seems "blocked" to give this structure to something that is battle rap or battle hip hop
Dont be cautious about adding something quite different from what you feel is your "genre" youre going for.
If the prompts are mostly hip-hop dark ghetto rap, add opera, or bollywood or isacanthemia at the end of the prompts to send it slightly off from the repetitive direction its been going.
Dont forget, you csn get some interesting things from "free interpretation" and "improvisation" as well.
Use [interlude] or even add [interlude] after [introduction] . If any vocals, we can crop off the beginning or fade in over it as an option.
@shy badge
This is better for this channel any way.
Thanks again Hep.
What that is folks is a kinda master prompt for chatbots. Create phonetics as well. Create a lovely breathable track. Minimal ai noise and minimal ai artifacts. You can also just tell the chat bot to replace the styles, lyrics etc to your taste instead of answering the questions.
I wanted to share the whole thing here and in general chat... Too long 😆
The random vocalisations in the intro is a bit of a quirk of v5, it's not restricted to battle rap.
I'm not aware of a reliable way to completely avoid it as of yet, but it's usually fairly straightforward to remove in the Editor/Studio if you split the vocal/instrumental tracks apart.
I have a payment problem and I need staff. #archived-ticket-help
Anybody has issues with match lyrics to audio in the new editor? It seems fine in the legacy version, but the part i need to change in the song (the editor doesnt recognize that part) is less than 6s. I have that problem a lot. Nothing seems to work for the button not to be shown grey, its just random that I keep changing back and forth between two versions and 1 percent of the tries the button suddenly becomes clickable.
Good that you already posted there and also already sent help to billing@suno.com
Now, just wait for the reply.
The problem I sent and I've been waiting for a response for over a few days and it's really bugging me.
Yeah - unfortunately it does take a while - support might take weeks but hopefully billing issue is sooner.
Hi- new suno.com user (pro paid plan). I ‘published’ one song, when I click 3 dots and scroll to both ‘share’ and ‘permissions’ I can’t see the option to make my song ‘public’, only the ability to share it as a link. I emailed suno support but have heard nothing. Any suggestions? Thank you!
I think when you published it - it'd become public and will be visible on your Suno profile.
You can still share its link.
According to suno’s own help center, when you publish a song it is not visible by default, new songs are set to ‘link’ only. So when you click the three dots by the song, and go to ‘visibility’, you should see the option to toggle to ‘public’ rather than ‘link. I do not see the ‘public’ option, and my prior research says other users have had exactly the same issue
Yeah app doesn't have that toggle.
Just publish or unpublish
Ok thx; I’m guessing I just have to wait for a response from suno support
The layout changed, but publish makes it public. As the word already means public. It used to be with the visibity options.
Ok, thx- I’ll ask a friend to try and see if they can find my published song without having a link to it
im attempting to make an album art that might be a bit risque but I wanted to get it passed by suno people first before just dumping it on to an entire album, any way to reach out about that here?
in your profile they should be able to click show all and see all your published songs
Thx, appreciate it
I've published a few in my free time just seeing what I can coax out of it. Like making V4 flex into sounding more like V5 😅 but I'm not sure how "easy" it is to locate newer songs. Especially newer low traffic ones.
My V4s can stay coherent for 4 minutes straight. Repeating the chorus with the same emotional strength
I’m assuming if I tell a friend my suno user name and my song name they’ll be able to find it, if in fact ‘publishing’ it now makes it public
Want to hear my tracks show what AI engineering does to V4?
Gotta use Audomee
I’m creating tracks from my piano and vocal demos as upload. I had a big problem when doing the cover thing with overplaying after like the first chorus. The drums doing 100 fills and all instruments and vocals go crazy. My music is in the soul rnb kinda style. Is it the advanced settings that I have to change to make Suno to create a track without it just going crazy after a while?
Seeing AI drift. It'll go nuts on unclear instructions. But it's hard to explain without going into engineer jargon 😅
So, what can you do with v5 then? 🤗
If you want to hear my v4 I'll dm it. I'm still on cool down in showcase
😅
No, I mean, what your AI engineering does to v5(!)
Just sharper sounding.
Sharp is good! 😁 Can you share something?
I'll dm one of my tracks. Dont think I can plop it in here 🤣
Got 50 or so minutes left for showcase
I don't know why you shouldn't be able to plop stuff in here... as long as it serves the purpose
Here's my V4:
Ok, it's v4.5-all ... what did you engineer to improve beyond its known abilities?
Hard to actually describe the mindset behind the scaffolding. What I do is professional AI LLM engineering. Suno is a simpler work project for me. But yeah I pushed V4 hard. But trying to explain is me turning into a professor I'm not quite awake enough anyhow to switch that mode on.
Reason why it's simpler is Suno isn't true LLM.
I see. Suno is only part LLM, though.
haha, agreed
well, not necessarily simpler, though
Well for me 😅
This is a lot less complicated than making a choose your own adventure AI Which I've done
ok, may be simpler, but less predictable?
Rhythm is definitely a bit random but I can lock tune. Here listen to this Nightwish one 
I don't know Nightwish so well, but I guess this is the original tune, but different lyrics?
It's just sharing a genre nothing uses their actual track. Nightwish I'm mentioning because they use a lot of orchestra
Ok, what do you mean with "lock tune" then?
Lock rhythm where it doesn't drift genre mid song or lose the chorus
Did not happen very often to me, except when fusing genres and stuff, but then it is expected.
It can happen badly on certain genres
But even then, there can be a nice back and forth and intertwining, like in this Metal-Reggae:
https://suno.com/s/scZ2nVbYI7kiaQgj
Definitely hear it drifting occasionally. But as an Engineer I can tell you Duets are basically broken on current tech. At least true control on it
Until we have multi agent
I know - we all found out pretty fast ^^
Just checking that youve seen the Publish button is on the main menu page? Any other "sharing" is not the same as Publishing (or going "public" ).
Check your own profile page. It will show only your songs which you've made public. On my page I have/see one Hook, one Persona and one song which I've made public. That will be what everyone else sees as well.
That seems to be the case on my phone app (only see the one song I published) but on desktop pc I see every song I created even if not published. I think my ‘issue’ is that the instructions I was using for making a song public on pc were generated by ChatGPT, which gave me instructions for making public that are no longer valid after apparent suno updates. The song I did publish apparently became public when I clicked publish, as I have confirmation that others can now find and listen to it. Thanks to all for your suggestions👍
Mind that people can only see your profile page, not your workspace(s).
Thx, I’m so new at this I didn’t even grasp or consider the difference.
they should probably adjust the way public v. published is referred, it makes it feel as if it isn't synonymous, eg. library
Hello everyone. I just renewed my account to PRO from the phone app and through google store. I got credits, but then my account switched to FREE...messaged tech support, but no response...
So you already sent email to support
Then just be patience and wait for the actual response.
Visit #archived-ticket-help if need a follow up.
does it show free on both app and browser? wonder if there is a similar disconnect as the other user a week ago
Yes, it is free everywhere 🙁 but if check the plan it shows PRO...
probably won't fix it, but, open the app info page, storage, and clear storage (not cache)
you'll have to relog in
but that should pull all data and completely refresh it, just a step, but, i don't think it'll fix it, but won't hurt to try
Thanks. Tried, nothing :/
