#📚┃suno-school
1 messages · Page 189 of 1
Hello, I would like to know why I can't use the persona I have in my favorites?
Did you create the persona first?
funny you should post that. Just yesterday I figured out doing all my prompts in the lyrics section gets better results!
analyzing.. looks super accurate to the art direction.
right down to the last Shakuhaci note over the ending.
its bugged. open song in editor, save as new cong (copies) create persona with the copy
Still can't get the stops thing to connect an extension to the full song. I click Get Full Song and I get "An Error Occurred Generating a clip." SMH
SUPER HELPFUL
Still experimenting. can be hit-miss like anything in Suno.
https://suno.com/s/SzQLMDaHU966bf99
Trying this question again in case anyone knows: all of a sudden I can't seem to generate with a Persona and Inspirations at the same time. am I the only one?
it worked all afternoon yesterday and then stopped at night. Can click the Create button but nothing happens
I have an issue where one of my V5 projects is stuck in the V5 Preview mode and keeps saying "Unlocking". I've tried logging out and back in after clearing my cache to be sure it wasn't a local caching issue. Has anyone else experienced this problem?
tested. same. pretty confident i used them together yesterday as well.
No Styles: Country Rock https://suno.com/s/NtEx7icTw4EdKjpf
No Styles: Southern Rock Steel Guitar added https://suno.com/s/cAm1FuSB2VC2ivkf
Drums dropped out too. [Finale: drums drop out; acoustic guitar and steel fade together; electric guitar bends final high note into sunset air]
I have conquered Suno! ❤️
So is no one else having the same problem with song extensions?
saw someone saying had issue with it earlier. they mentioned v3.5 involved tho. was that you, or?
Hang on and let me check. I don't know how old the song was. I was testing on an older song
half way through testing 3.5. v5 worked fine
R.I.P It's a song from 3.5… Thanks @frozen dagger I didn't see when someone said that
Brackets seem to appreciate semicolons [Final Refrain: double-time drums; soaring lead guitar; full choir harmonies]
just like any coding language
firmer separation than a comma, i think you're correct.
im listenting and waiting on gens to finish to see if it works. but 3.5 extendsion to 3.5 worked (this was on gens i just did not older 3.5 or any earlier models. could dig one out to test that but so far loks like you just cant go from a lower model up to 5, but you can go up to 4.5+
I was testing extending a song to see if vocals changed because I hadn't done that before and my friend really likes a song of mine, but wanted it to be longer. So I wanted to see if I could extend it without goofing the vocals because I just wasn't sure. I'm overly concerned or whatever by some stupid hardwired nature B.S. in my brain…
it usually keeps mostly consistent, at least to start with, but it can swap in totally new voices or distort it. using a persona with a vocal stem from the song you are extending can help keep it consistent.
only 20 credits left.. 😢
When I get to Studio I will be able to save like say Chorus 2 I liked on a song and build around it?
Or you like a solo and you can keep just that section?
you can try splicing them yourself (with some success) or place the sections where you want and then cover it to create a new song that will follow your little sonic blueprint of the song... with some creative liberties to try and make something that may be pretty choppy in editing into something not so bad.
hey i was playing with Suno the last few days and sometimes it just cant pronounce some words is there a known fix for that?
phonetic spelling pretty much usually is the fix. can fix the lyrics in Displayed lyrics later. some issues with dialects and pronunciations is same thing relying on phonetic spellings.
For this song for example I had to use Eee-ry-na
https://suno.com/song/c0912835-81a3-4e91-9d5d-41c7703c0033
ok to be honest eng is not my 1st language so i am a bit cofused atm what you mean as some points but for example my current song got the word ARKFORGE FLOW exactly written like that but it sounds like ARK AGE FLOW and at the start i wort Arc correctly then it just sounded like mumbeling
try hyphenating the syllables to get it. "Ark-forge Flow"
or a space arc forge flow
ok thx will try
other thing is there a way to edit a alredy done song since i like one a lot but he got that spelling mistake
edit the song lyrics displayed? Song Details -> Edit Displayed Lyrics.
If need to fix a word that is sang in the lyrics thats different can of worms.
ok thx
Have you tried email billing@suno.com ?
Yes, but no reply
They tried to take a payment, a day before payday
https://suno.com/song/4edac43c-092b-4d3e-9b7c-7c5a659c4ced i did try both of your sugjestions and i sadly had the same results could it be becouse of the gere of music i used ?
They will - I guess just wait a bit - then visit #archived-ticket-help for follow up.
Ok so I have a song. Got the result I was looking for. Now Ive split out the stems. On the drum stem there are parts that are glitchy. Dropped volume, sound artifact etc. I want to tell Suno to replay the track. The only options I can find just replay it a diffrrent way. Is there something I am missing as to how to prmpt a new copy exactly the same just without the glitches?
"AHRK-forj floh" tested , but with v5 i hit too quick but covered wwith your style sounds better
ab-so-free-keen-loot-lee
Maybe try in Studio - regen that glitchy section of the drum? create a new track - then select the area, select the instrument (drum) and recreate...
THX for the help ^^
yeah tried that. It basically creats something new which most of the time doesnt fit the song anymore
can fix spelling after. phonetics typing for the sound and pronunciation (or form; like hyphens - can create Staccato phrasing, or em dash — can create pauses or hold notes (its contextual there).
or you want it sang with some funk, "I'm going cra-a-a-a-a-a-z-y-y"
It's not perfect - yeah - but that might be the only way in Suno to fix that stem
Or you could try to Cover the song and that might give an exact same song but no issue on the drums?
Or replace the instrument off line in your DAW if you can as last resort.
Yeah Im tryin to basically slpit out the drums and then bring it into my DAW and use Superior Drummer on it. Problem is the Stem is so crap quality the midi is terrible. To make it work basically have to write the midi on my own. And I am not so hot at that
The question is this: Why has Suno been making songs and music for me like Arabic metal for a long time even though I have marked Celtic folk, Nordic Folk or Celtic folk metal or nordic folk metal?🥺
Hey, after the v4.5-all update my paid account got downgraded to Basic and most of my songs are gone. I emailed support 12 hours ago and haven’t heard back — any way to escalate this?
would you be able to share one of your songs doing this so I can see your prompts?
i deleted some of my songs already 🙁 sorry
thought this might be the case. ok.
https://suno.com/song/cd2206b2-b215-485f-8ace-a5935f4a20b3 something that but this is not cetlic or nordic. Melody line reminds me of arabic quite a lot
Backing to free version. Does anyone know how long do I need to wait after cancelled subscription? Showing "Cancel Scheduled" for long time now
What do you think of this?
https://suno.com/s/1zUUB6fkRduo0HQh
Be sure to listen to it. it gets heavier.
Here is the other one that gen'ed. https://suno.com/s/X5pBe257WdEswuqz
Def Celtic!
and Heaaaaaa --- vvvvvvv ---- yyyyy!
3.5 is dead.
This exactly Celtic and a litle bit nordic. I imagine something like this but with female raspy and alto voice. And these Lyrics I wrote myself abot 21 years ago and original text is in Estonian by me.
Use this template and experiement. Hit it up! sky the limit.
What this :zap means exactly?
Thank you 🙂 And You are German or English?
American on one side to 1600's. My grandfather left Germany in 1936. but do not speak it.
Ah i see. 🙂
See if you can get it to be female.
Can force with button.
I got the German "inventor genes" lol
3/4 is usually the time signature of a waltz. 🙂
mix it up. its a sad song. 😢
https://suno.com/s/s29MeoyqG3rp7bc3
Actually, I guess I should mention that these words are copyrighted in Estonian 😉 Because they have been published as poems in a joint poetry collection
Ditch your style prompt.
Causes confusion.
Sorry what means "ditch"?
temporarily abandon. the songs I showed you do not use them.
experiment.
Ok thank you 🙂
Experiments are still allowed, because it also gives me new ideas on how to do things 🙂
want to see how fast I can make an atmospheric nu-metal rap song?
when will this Woodhouse APP appear in this conversation?
I blinked, do it again.
Oh okay. IDK how to do all of that. The vocal STEM with the persona I mean.
Sorry. Was getting ready for an appointment that I am now at and waiting to begin
I'm trying to find a completely new style that consists of Metal, croove,rap. hip-hop, phsydelic, industrial metal, Celtic or nordic folk or maybe rock 🙂 but Ultimately, it still sounds more like NDH 😄
no worries that why we have discord and notifications. Just "get stems", on the vocal stem create persona from it. (may have to create a copy cus it buggin' rn; open in editor save new song then can make persona if it gives a prob on the vocal stem you separate.) now the persona can be used in addition to your extend to help try and keep same vocal sound
So... Copy the song or what am I doing?
Sorry. I'm not awake enough yet, plus I'm at an appointment. So my brain… 😖
I have a copy made with the same timbre, usually if I haven't changed the Style box. I haven't changed anything from the ligatures there.
Nu-Metal-Rap-Atmoshperic
https://suno.com/s/kloX1vZimSOkrtSA
I like this heavy groove more: https://suno.com/s/G2zw6wLSxDeDxgGV
How do I open in editor and save as new song? Ugh. This headache isn't helping…
first go to get stems. separate your vocal and instrumental. Then find your vocal stem in your workspace. (may need to uncheck "hide stems" in your filters, iif you dont see it) no right click your vocal stem and ccreate persona.
if it is greyed out then under "Remix/Edit" choose "Open in Editor".
this is only if create persona doesnt show upp for you on your vocal stem.
In the editor you "Save as new song". This cretes a copy with no changes. Go to that new song now and should be able to create a persona
after you have that vocal persona made from your song, you could use it to cover, or extend the song it came from. this how you can help keep the vocal the same if its going off the rails
I'm taking screenshots of the tips so I don't forget because my appointment is about to begin and it's a bit of a big deal and will keep my brain occupied for a while today.
Bit of knee repair…
So I have trained my AI Agent to create backing tracks as well today...
Also... realized another benefit to FULL lyric control...
When I publish my songs on Suno I can edit the lyrics. so my prompts wont be public.
Backing tracks are a nice place to start Lyrics from sometimes I noticed.
Or Instrumentals
Hehe fitting I open Studio for the first time while working on my song Studio Door 😄
What’s the suno link to fun 2 rhyme part 10 million
Do you know if there's any way to make that change without studio?
Adding specific instrument tracks is a Studio feature. The only thing I can think of outside of Studio is to add the flute parts yourself in your DAW, then upload the new song and have Suno do a Cover of it.
You can also try to prompt it to add them behind certain lyrics with bracket notation. I have no idea if this will work though. Just an idea. Suno can be quite flexible/experimental with bracket prompts.
Unfortunately no; not that I'm aware of on Suno.
On a Pro subscription plan you can try highlighting part of your song in Editor
Then adding [Instrumental] in your lyrics panel on the left approximately where you want them to be
And generating new samples
Something like that #🗣️┃chill-chat message #🗣️┃chill-chat message
We cant search Lyrics to find it by that means. User name? perhaps to then see if its public on your page.
hi guys im a little confused about some ownership stuff. if i am using the free plan but then have a good song, can i keep the instrumental and change a bunch of the lyrics to then make it my own?
With free plan - Suno owns everything it generated.
You can use the songs - just can't monetize and call any parts your own.
See #1294423389515284552
what if i make a song with my own lyrics but the instrumental is sunos? do i have the rights or no
if you make the song during active subscription then yes. If you made your songs while subscribed to a Pro or Premier plan, those songs are covered by a commercial use license. This allows you to distribute songs to Spotify, Apple Music, etc, using the distributor of your choice. This also allows you to use the songs in content you post to YouTube, TikTok, etc. You don't have to attribute the songs to Suno, but we wouldn't mind if you let your audience know we helped you create your masterpiece!
@broken veldt
uh...that wasnt hidden, just a hyperlink
hey question. how would i get suno to make something similar to Mr. Robot soundtrack? i want a song similar to this https://www.youtube.com/watch?v=sKUUeVK5iMo song which was from the Mr. Robot score
FYC: Outstanding Music Composition for a Limited or Anthology Series, Movie, or Special (Original Dramatic Score) “American Horror Story: Delicate” Voting is open June 13-24! #music #composer #fyc #ahs #emmys
You can try to upload songs to use as inspiration, but they must adhere to copyright and might not be allowed by Suno...
Another way would be to get structure and style of the song you want to recreate and type it out as prompts into lyrics using [ and ], and prompt styles too
You always own your lyrics if you wrote them.
Everything else including the song that generated with your lyrics are owned by Suno (on free plan)
With regards to the shimmer heavy songs from v.3.5, I have a theory, that it might be possible to use Suno to correct it through it's remix features (cover and persona), I'm personally trying to restore a few masterpieces (subjective), but there are some bumps and obstacles on this road - anyone who has experience with this? My theory is that using v.5.0 (or perhaps 4.5) a persona base on the track, and the track as source audio + possibly the original lyrics to help vocal adherence, then set Weirdness: 0; Style Influence: 0; Audio Influence 100. Also the result is quite different if you use the feature natively (create+remix) or from the Studio (remix).
That’s what I’m doing at the moment but my question was how does one get the melody to be consistent? I’ve tried using the same amount syllables so that the system would be prompted to try follow the melody exactly but it’s very hit and miss… I guess I just gotta keep generating until it spits out the desired result?
anybody experienced missing all the projects in the account?
@winged gale You can manipulate BPM in Editor, on the bottom row you can find cog wheel
Usually when this happens, a person uses sign up instead of sign in accidentally (which might cause your original login to get reassigned) or you are logged into another account.
- Check to see if you are logged onto another account
The rarer situation is a bug in the backend or server issues that delays/interrupts/breaks the API call related to your creation info (your projects are safe, just does not appear until the issue is fixed). This usually results in everyone being affected so you would see panic/chaos in the server.
My nightmare whas If have lost all tracks but I feel safe now that you mentioned ,actually it logged me out randomly, only I have 1 account and it changed my username under the same subscription when I logged back in and of course...non of the projects
Now I have a clear understanding so I made a duplicate under 1 account basically
As part of this test, I took lyrics and just created a new song with the lyrics, but with NO style prompt, NO exclude, sliders all at 0.
The reason I did this, that during my attempts and tests to restore the track (generate a track close to original as possible), but even keeping the sliders at various levels, it kept having metal genres bleeding through (especially industrial and NDH at some points), either quite heavily as a base, all down to being more a subtle base. The original track to be restored, has those exact tags in it's style prompt, but not really part of what was generated.
The original style prompt: "Electronic, Pop, Industrial Metal, Synthwave, EDM, NDH Mellow, Big Beat, and epic cinematic atmospheric and eerie elements, Powerful Male Vocalist"
My theory is this has more to do with what Suno "thinks" it's genre and style should be based purely on the lyrics, when it has no styles to go on, but when there is latent imprinting (as when remixing), it will tie this to most likely genre based on the lyrics - hence the test.
Test - 1st iteration: https://suno.com/song/7b25e9b5-5aa8-400a-b6c7-b42690cd6b45
Test - 2nd iteration: https://suno.com/song/37f8c10a-19dc-4e52-86b7-e910847a351e
The original: https://suno.com/song/c7907877-25a5-4b68-aa40-b1c82eed67ec
.. I can provide some of the other tests where you have the Industrial Metal/NDH bleeding through / being it's base, despite it's even attempted to be excluded.
Further testing, 1st iteration and 2nd iteration test, was in the same workspace as the rest of the generated tracks in this project, so to rule out that workspace somehow can affect, I created a new workspace - they also sound to be in a similar style and genre as 1st and 2nd iterations. It kind of further proves my theory.
1st iteration in new workspace: https://suno.com/song/5eea8859-beaa-4940-a4a2-0b430d477e62
2nd iteration in new workspace: https://suno.com/song/62f59dc6-ba29-4e54-afbe-c1b71b23a2aa
Another test, this time using the Suno Studio (remix/cover):
1st: https://suno.com/song/81e9005f-c558-4738-9994-a75ff7ea07cd
2nd: https://suno.com/song/6ed7d4c1-4393-48e3-8c20-94d8774effcb
..still Weirdness: 0; Style Influence: 0; Audio Influence: 0.
Kinda, it's more like you got a new credit card when you tried to renew your old card but accidentally signed up for a new one.
There are two causes for this:
1a. You actually logged onto the wrong account A lot of people assumed their Suno account was assigned to email X but it was email Y. You can verify if the email is correct by searching for emails from Suno to you
1b. Similarly, you may have logged on using the wrong SSO, though this is an assumption I did not test. This typically shouldn't be the case but there is a chance that signing in through Google when you actually created your original account through Discord
- The system may have accidentally overwrote your original account ID with a new one (as I mentioned before). For this situation, you need to email Support to get this corrected. I do not know the current turnaround time but it is probably best to work in an alt account until Support corrects you.
Just in case you were unaware, setting sliders to 0% or 100% can result in very weird/broken results. Unless it is intended, you may want to make your values range in more of the 20–80% region
I want to give you a special thanks for the help ,sometimes we get senior vibes jajajajajajaja you helped thru with the vibrant answer
I whas able to get it back
Ah great to hear 🙂
I've tried a variety of slider settings including lower values, with and without prose style prompts that are close to identical to how to how the song sounds, with and without exclude list. But what I'm attempting, is both to try generate a song that is as close to the original, but also a methodology, that others can follow.
What I've so far uncovered, that the combination of things that causes the end result:
- The imprinting in v.3.5 track, carries over differently in v.5.0.
- V.5.0 being better at capturing the styles, genres, instruments, etc.
- Lyrics have latent style and genre gravitation, which Suno guides towards the styles.
- The more cues towards a style and genre, even without Style Prompt and Exclude list, will make it lean harder into it.
- IF the underlying imprinting is present, and the lyrics lean this way too, stronger gravitation towards the so far established genre, style, etc.
- IF sliders facilitate stronger Audio Influence (Audio reference approximation), it will lean into the genres, styles, the lyrics leans towards - despite what the USER THINKS OR EXPECTS based on the original.
This is just an attempt to fix the synth that is the one with the most shimmer:
This is what I got: https://suno.com/song/a07d243a-dda1-4453-8f13-78a8febed39c
Still Weirdness; 0: Style Influence: 0; Audio Influence: 0.
It's not that it doesn't sound good, it sounds amazing - but it doesn't sound like the original - which is the intention that it does.
hi anyone knows, why deezer is detecting all songs from suno as AI Songs?
because they are AI?
You do realize that Suno is an AI that create AI Songs, right?
yes but its crazy that they brand the songs in the stores automatically as AI
you mean every song, not just suno's? Maybe they get so many AI entries it just made sense to toggle it by default
or they just have really good AI detection
either way, you can't really complain your AI song is being flagged as AI
yes it is branding every user which is uploading his song to this plattform
i havent seen this before on any other platform like apple music, spotify or youtube music
suno is making a watermark on each song you have made
and deezer can detect its origin
that text provides no evidence of a watermark
deezer probably works like any other AI analyzer and trains on songs generated by the platform to detect patterns
They have admitted that they embed a digital watermark, which is as much to protect them as to protect the owner of the track.
However, they're very likely doing that as well.
it's a proprietary system I doubt deezer has access to
All they need an algorithm that can detect it.
It's good, though. I've always said AI generated material should have watermarking
Embrace it - there's no shame in it.
Enjoy the music!
I'm just pointing it out.
Those who master AI Music, will be the future leading music artist.
Though you can always make manual changes to your tracks in a DAW, which might change what detectors say.
You use this free tool. https://www.submithub.com/ai-song-checker
Actually, there is one producer on youtube that made a video essay on how he and another guy figured out how to recognice suno. Dunno if it still works for v5 though
not by training an ai
nvm!
double checked, and they did use ai
lmfao
Man this brings me back
to when shimmer was the actual devil
Studio seems to drain credits fast.. went from 10k to 9750 in a blink somewhere..
I was wondering why studio was a premier feature and not pro. Maybe that's why - so you have more credits to play with xD
Yes, and rewriting just like a bass line I think.. but still a little lost in Studio... played a tiny bit last night.
I had a really dirty guitar, gonna see if i can get it to do clean strums.
ye I noticed some of the instruments are a bit dirty with v5. sometimes you get a clean mix, but I found it rare
literally learning how to use MIDI in daw just so I can use better quality instruments but the same melody
I Have a midi piano that I've never really used for its intended purpose anyway
that's why I was realllly looking forward to studio, to get those MIDI exports. how dare they lock it behind premier >:(
creds to suno for making a product so good, that it makes me learn completely new software just to make up for what suno gets wrong
if you don't have premiere I can export the midi's for you
send me the wav when ur happy with the end song, I'll upload it to suno and download the midi's
then delete the upload
how many credits does it cost you to export MIDI?
ehhhh ty, but I don't want to waste your credits. If I could send you credits I'd definitely feel better about it
lol thanks, I'll keep it in mind. appreciate it!
@wise kindle Get NeuralNote - it's free.
It's basically a program that let's you transform audio to midi.
What I recommend doing
- Get the stems - You can use this tool: https://ultimatevocalremover.com/ - Free and exceptionally good.
- Use the instrument stem with NeuralNote
- Import the midi to DAW of your choice.
Download here: http://www.github.com/DamRsn/NeuralNote
Tutorial here: https://www.youtube.com/watch?v=QccK8jkQx0Y
Yesterday I've cancelled my subscription. Still showing "Cancel Scheduled". What that's mean? I wanna cancel not waiting for a meeting to convince me to not cancel.
Because of this "Schedule" I'm not able even to use v4.5-all . Anyone please?
Need some help with cover / extend of existing songs. I uploaded a couple of instrumental pieces from my youth (1988) into suno. Covers are sounding great, however I'm unable to cover the whole length. It picks a specific section and goes to about a third of the whole length in covering. Any ideas? The uploaded songs are naturally without structure tags, just [Instrumental]. They are generally 5-7 minutes long.
I read about extending the first fragment which resulted from the cover, but (how) will that extend from the bits of the original covered song?
You'll want to get a ticket in #support-guide . This tech-prompt-talk channel can't really help with that unfortunately.
I did, also I've email...no answer do far 🤷♂️
i made a song wity my own lyrics from last year but the instrumental is sunos but when i get the subscription plan i feel like its gonna be seen as sunos work and that i wont be able to use my lyrics anymore. im really confused and the only proof ㅑ have of it being my own is notes but since I put the instrumental now in it the old date isnt there. what do i do
What happened to the original? That date won't change. Is it in your trash bin?
no i downloaded it and put it on the old one instead of saving it just incase i cant log back in to suno but it changed it to today😭😭
nvm i found it i just sorted my notes by date created and it shows the original date 🙌🙌
If it can be detected, then it can be also removed.
Has anyone ever attempted a bueling banjo type of arrangement?
Can it be done?
What other instruments would you consider using ?
Maybe with something like this and a strong reference inside the styles section.
[Instrumental Duet: high-energy dueling banjos-style call-and-response ]
If the first time those lyrics were entered into Suno was while you were subscribed, they remain your own.
Suno retains ownership of the original instrumentation which you're using as music, as you created it while on a free plan, so the combined song you've made is a derivative work of something Suno owns. They do allow you to re-use it, but only on a non-commercial basis.
For clarity, you also don't have full ownership over the new song - in effect, ownership is split between you and Suno. So unless you agree otherwise with them, you can only use the new song on a non-commercial basis as well.
i need help
when i want to extend a song, cuno always only let me from 5:40....but i want to extend from 7:20 time....whats the matter with that? i dont understand this....a few songs before suno let me do extend song from 7-8 minutes...why not now? i want to finish a song but suno only repeats almsot the whole song....
just drag the extend bar c:
i did....suno just do as i said....the drag bar doesnt work for the song....
and writing the timestamp manually?
did it....doesnt work...i tried everything but suno extends all the time for the song at 5:40...
In which case, I'd recommend cropping out a bit of the beginning. Find a place where the transition is easily patchable later on. Then extend from your new section (the one that doesn't include the beginning).
i decided to let the song as it be now....i tried to extend it the last hour....
the song is 8:07 minutes long....wanted to have the last sentences a bit different....but its okay...i will leave it as it is
...I....see...sounds....good....
Why Suno not greate cover of my before greated Music?
You mean create, not greate, right?
Create button in Suno I mean, Suno not generate covers.
Suno "Create" button is broken?
Works for me, without any problem.
Yes now seem works forme too, but i can not upload my mixes because 2 or 3 time i got massed it match with anither existing artwork. Question is How if i made this about 7 years ago wand i used tablet music studio+ FL studio+ magic music studio greated mix and Suno say that thi smatch with another art work.
Can you posibly use some kind of translator, google or any other? My English is also really bad, but yours is really dificult to understand.
Create does not work its scrolling and scrolling but not generate new covers
It takes some time to generate.
I have nbeen waiting about 5 minuts now but still nothing
Could sometimes even take 6 minutes.
i dont know , before did once in about 30 second 2 covers.
Sometimes it takes just much longer.
And one thing is alittle bit strange there in suno all music what i have did before and upload thetere Suno say that " A High-energy industral ..." but actualy this music have not High-energy, them have been atcually calm and pretty slow music mixses.
Has anyone had success getting suno to have two distinct vocalists? Specifically I want to make a heavy metal x rap song but it always just makes one vocalist do the rapping and the screaming even if I explicitly state in the prompt I want two vocalist.
Actually, I would also like to know what trick is used to do this. So far, it has been completely random if I have checked the style box, for example, female voice, male voice.
add specific prompts in the lyricsection... for example [male rap vocals]
btw... what the heck is going on with the high frequencies of the outputs in the last days??? hihats and ride cymbals are sounding metallic as f**k 🙁
It's weird, but for me... for example, [male rap vocals]
don't work. 🙁
well... it's still the "suno roulette" 🙂
And what's even weirder is that when I had the free version, checking the "female voice/vocal, male voice/vocal" box in the style worked wonderfully. 😂
every model has his own pro or contra 😉
And suno generated a song in the free version where the woman sang parts of the verse and the choruses were a complete duet, just perfect.
I've also tried to achieve this in the pro version with Sunoga, but so far I haven't been successful.🥺
well.. v5 creates nice polyphonic vocalparts for me 🙂
You mean more then 1 voice singing at the same time?
The song editor is an absolute JUNK. The versions that platforms comes up with never fir the rest of the song.
How you do that, maybe you say away this trick? 🙂
let me search a track out of these thousands of outputs.. will send you an audio via DM then... may take some minuntes to find it..
but for that you must add me your friendlist, because once you can't take me DM. I am set this that, sorry 🙂
I signed up and paid , I signed in but the app is not responding like a piad member. anyone got this issue too?
you have to add me coz you've disabled friend requests 😉
hm strange, ok i send you request.
ok...added.. now i have to find the song i want to show you 🙂
Now you are there. but maybe it was something else not me.
need to switch to my computer... hold the line.. it will take some time 🙂
One way possible is to have a persona set up for each vocal. Then extend while adding the chosen vocalist persona on every segment that there would be a change.
For new forms of payment, the funds are often reviewed for anything suspect. That can delay things, even a few days. That could be why your subscription hasn't activated yet.
Ah ok I did something similar to this but for a completely different use. Dont know why I didn’t think of trying this.
I removed a verse to get rid of weird instrumentation and then made a new cover of the shorted song to get back the verse.
I found a place on how to create this Persona in Suno, but the problem is, can I use the voices of the existing song characters or do I have to create it entirely myself somewhere else?
@steady kelp : are there known issues about shimmering on high freq. with drum outputs (v5) ??? Hihats, Ride cymbals and stuff like that are really bad round about 1 week... it's not on edm music, but on typical rock drums.. muddy, metallic and shimmering... i think your company should take a look into that 🙂
I got a Faith No More type Piano ending (postlude(?)/outro), I want to extend it like another minute with just Piano - but it always brings in the other instruments.
evxtend with [piano solo] in your lyrics prompt? could use a clip of the piano as persona, or duplicate the segment and then try to recreate so it will be one longer piano play instead of a duplicated segment
Some good ideas, let me get a clip and Persona it first.
faith no more type piano is the good idea. Mr Bumble too
Bungle?
ya thats one way. also the third option highlight the section you like, duplicate, highlight the duplicate, then recreate. it should try to keep with what its has (the copy) but try to creata cariation that blends with the music before and after.
Why when i extend a song i'm not getting the whole song + the extended part? I'm only getting the new part.
Will be a Get Full Song button on hover to join them @ivory peak
Nothing happens when i click on it.
double check your "my workspace" folder
Yikes. Created 5 of them. There should be an indication or popup that it actually does something.
remasters and "get full song" appear in my workspace, its not always convenient
So to download song on IPad (step 1) click three dots (step 2) hit download (step 3) hit the mp3 (step 4) save to files ( step 5) hit open in files ( step 6) select the folder you want to save your file in.
Were some features avaible in 3.5 separated in 4.5 and made Pro only?
Is there any possibility of having an easier way of downloading? I don’t know everything in your workspace in a zip file could be a possibility . ? it took me close to eight hours just to download 500 songs.
Such feature should defenetly be avaible for Pro users. I'm not a pro user myself, but I'd like to see a feature like this for pro users. At the very least, it should be possible to download entire playlists.
When you’re a free user and have access to the V5 preview, you need to pay to get the full version. If I subscribe, do I get commercial rights for the music?
Hello, no, songs made before subscription do not get retroactive licensing
https://help.suno.com/en/articles/2425729
Subscribing to a Pro or Premier plan does not give you retroactive commercial use licensing for the songs made with a free plan by default. Songs ma...
itt doen't reply to my question because the preview is not the full song, it's only 1min, and it's written upgrade for the full song.
So that case is not in it
Song was technically created on a free plan, even if you extend it to full version, original song was created on a free plan
Here’s a natural English version of your sentence:
Because I don’t understand why they give just one minute and then say you have to subscribe to get the full music, but even then, you don’t get the rights. It doesn’t make any sense. They should just provide the full music.
As for "Upgrade for full song", that is because v5 model is only available to Pro and Premiere subscribers
So i got to upgrad to get the full song, but it's still not mine ?
I can't see the logic behind it.
read the terms of use.. 🙂
From the article I linked above:
Songs made with Suno's Basic plan can only be used for non-commercial purposes.
It does not state anything about model that was used for song creation, 1 minute preview songs are made to promote their newest model (I would assume)
So, in basic terms:
Song made on free plan - belongs to Suno (with exceptions)
Song made during subscription - your forever even if subscription ends
i did find the V5 Preview on it bro. I'm saying this because usually we got the whole song, and we know that it doesn't belong to us. But the preview gives One minute and say upgrade for the full music. This isn't the same case
Hope this logic makes more sense
From my experience upgrading from Free to Pro, songs that were 1 minute preview, were already full songs I did not need to extend (if I remember correctly)
So basically, what you’re telling me is that to encourage free users, you let them listen to a one-minute V5 generation of their music, and if they like it, they have to pay to hear the full track....?
free plan = no commercial usage
Yes and no, because when you pay for Pro or Premiere subscription you get right to all songs you create forever, more tools to use, other models including v3.5 and with Premiere you get Studio access
Exactly, but since suno doesn't give you aces to pro features, that 1 minute sample is not really good.
Giving one minute preview is a hook to let you know how songs would work in v5 (newest model)
I know about Pro and First—I was a First user for two years. But I’m talking about this specific situation. Basically, you attract people by letting them listen to one minute of music, then say, “Pay if you want to hear more.” But this music will never actually belong to them. So the person pays because they like the music, but it will never be theirs. It’s this method that I find strange and unusual.
but if you wanna hear all the song you got to pay...but it won't be for you anyway..
But when you pay, you still can generate the songs and maybe even find something better? (which is usually the case)
And for those songs you will have rights forever even after subscription ends
Maybe but you gonna pay for something else then. But the reason, you paid, it's because it's written upgrade for the full music...
It is as simple as "Were you a subscriber at the moment, yes/no?", promotional models and other bits and bobs do not matter as stated in terms of service...
the rules are easy... free plan tracks = private... pro plan tracks = use it like you want
And you get a full song; they do not state "Upgrade to CLAIM this song"
That's what i means, that's the reason, i said it's unusual.
Maybe I do not get where the confusion is coming from
That’s what I’m saying—it’s that which I find strange. Because when you’re a Pro or First user, you actually get full ownership of the music after paying.
You get the full music after u paid, so during the fact you're a pro or premier
but it's still doesn't belong to you
No?
yeah because u got only the full music once u paid
no before
so when u get it, u're not a free user
That's the difference with the others that u get fully without paying
So, here are facts we know, okay?
- All songs created during Free plan - belong to Suno
- All songs created during paid subscription - belong to user
- When they give you v5 preview song, it states "Upgrade to HEAR full version"
- Nowhere it states that you will retroactively acquire rights to songs generated while you were on a Free plan
correct
You got the song when you hit "Generate/Create", not when you buy the upgrade
No it's written : " Upgrade for Full Version "
Yes, full version of a song that was create during your free plan
look
But if I pay for the full version, I only get it when I’m a Pro or First member—not as a free user.
But 1 min sample is not really a song, the real song will be created after the subscription.
Yes, click on the song title and look at the date:
If this date is when you had no active subscription - song belongs to Suno
This is what i'm saying. So why is it that if someone subscribes to get it, they don’t actually own it, even though they would get it as a Pro or First member while their Pro or First plan is active?
As I stated here #📚┃suno-school message
The full song generates on their end, you can only HEAR one minute of it
Well, as far as I remember when I upgraded
But for the sample you have defenetly no rights, but who cares about a sample, which is useles as it is.
But if you subscribed, it's not the sample but the full song. And if you like it, it doesn't belong to you. But you got it only when u where Pro or Premier.
Let me walk you through this one more time:
- You clicked on "Create" when you are on a Free plan
- Four songs were created, two with v4.5-all and two with v5
- Those songs created with v5 you can only listen for one minute
- All four songs belong to Suno, no matter the model that was used
Hope that makes it clear
You do not get retroactive licensing to songs when you buy a subscription, but you also do not lose rights to songs you made during subscription
It’s like going to a video game store — they let you try a game, and at the end they say, “Pay to play the rest.” But then, once you pay, they give you a completely different game.
But the song you have created, that are preview 1min. Once you paid, are they still 1min ? Or more?
none of the songs, no matter your subscription tier, "belong to you". you get commercial use license.
They are full songs, not 1 minute, you can only listen to 1 minute
I know, I got such samples even twice, but they were garbage, since they were 0% of what I wanted. Sure the music quality and the voice were beter than from free model, but since it was not what I wanted, it plaied no role.
But we are not talking about quality here...
Okay i get it. But i can tell you that, this is tricky. Anyway. Because you cannot tell that the person is not paying and thinking that he'll get th emusic
they are ai generated. there is no copyright laws currently in most parts of the world where you own them. seems like a lot of huff n puff over nada
The person who pays because of the message will think that the music will belong to them.
It is a bit tricky, I agree, but the logic in Terms of Service is very simple.
If you were on Free plan when you clicked "Create", no commercial rights.
If you clicked on "Create" during subscription, you have commercial rights.
Honestly, I did not think like that, because before using Suno - I read their ToS 😄
I know ToS are long and boring, but if you do not read them, you might end up like in South Park episode!
i know their TOS, that's the reason i ask
because for me
That's ain't logic to do that, it's opening a door
yeah but for logic, why are they paying ? to use V5 ? Or get music in V5...
Do you see the difference?
They are paying for a subscription, which will give them commercial rights to songs they create during it 😅
And with subscription you indeed will get access to v5 and ability to create and license music with v5
Just not those tracks made during Free plan as promotion
No for most those that will click for : " Upgrade for the full version "
Picture this: one day Suno decides to make v5 free, for a Christmas or New Years as a promotion
People create songs during that free period, they some people decide to buy subscription
It wouldn't matter that they NOW have subscription, songs made during promo were made on a Free plan
Because what they’ll be thinking is: “I paid to get access to this music.
Just to be clear added watermark, to avoid confusion for other users...
I would agree with you if the text on the button would be like this
But they give you full song, that was created by you on a Free plan, so...
Case closed 😛
only when u paid...and u get it on Pro or Premier plan 🤣
Eh, technicalities
Maybe better text would be "Upgrade to hear full song", you might suggest it on #💡┃suggestions-web!
Hope you have a good time of the day 🫶
But what about the lyrics — do they own those as well? Or is it just the music as it is, without the lyrics?
It depends, but as stated in this article:
https://help.suno.com/en/articles/2746945
If you wrote the lyrics for your song(s), you own those lyrics. Most copyright offices will allow you to register those lyrics on their own, and you may be able to use those lyrics to register your whole song as well.
Copyrights are complex, and vary by region! To ensure you have the most up-to-date information, we suggest reaching your local copyright office before...
To read more about all things related to copyright shmopyright, take a look at #1294423389515284552
thx
No problem!
any good prompts to get a synthetic/vocaloid sounding voice?
can do robotic, glitch vox, and anythign to decribe what kind you want, like down-pitched, time-stretched, etc. Can also down-pitch a vocal yourself and use it as a persona to create some weird stuff
yeah, but i don't have premium, so no persona for me
Can someone help me with two things? Cause it's doing my head in now. I've got a song I'm putting together, and told it to use my persona. Yet the damn thing won't use the voice of the persona. And also for some reason it's reading "live" as you "living" not live as in "We're live"
I will answer here too:
- In the advanced menu, use this slider to control how much of a persona/original audio you want
- Try using phonetic instead of an actual word, for example "laiv" instead of "live" (later you can fix lyrics to display normal words)
if you're using a persona, and also covering a song, or covering an audio file while using a persona, which thing is the audio influence slider controlling? the similarity to the original song, or the persona?\
This is really weird, woke up to a song I never wrote in my profile, must be a glitch, it’s someone else’s song….first two songs.
I have become a little confused (whom I kidding? a lot confused). I have a song, female vocal. In the second verse, I want the third line to be a male voice (a response). I thought I read a while ago how to do it, but that doesn't work (Put at the beginning of the male line: [MALE VOCAL] ). Any help would be appreciated
Does anyone have any idea how to force SUNO to add thunderstorm noises? I've tried multiple times and it just refuses.
I have a song that does this there's actually a male speaking several times throughout the song and suno just does not understand it. 🤷🏻
To clarify I have called it out with [male vocal] and then I've called out the next line with [female vocal] and back and forth and it still doesn't understand it.
Hopefully someone "smarter" than us can help 🙂
You should also add "duet" to styles and still it is not the most reliable thing
song card isnt showing up on right side
Duet to styles. Will try that one
Try to refresh the page with Ctrl+F5 to see if that will help
nope
Hmmm, what if you try to click between songs, click on the one above, then go back to the current one?
Also, might be worth a try to see if the same thing happens in another browser
Is it possible to slow a song down? I have the perfect result I wanted, except the song seems to rush through everything. I have the voice, the emotion, and the music itself, but it's as if it's singing it like a race not a song
^
If you are on a subscription, you can use Editor, right click on the song, "Remix/Edit->Open in Editor"
After it load, on the bottom you will see this cog wheel and a number, that is BPM
By adjusting it you can try to get results you are aiming for 🤗
That's kinda done what I need, but the main issue is it ends a verse then IMMEDIATELY jumps into the next like it's wanting to get straight to it, without taking a breath
Like it finishes verse 2 for example, and the second it finishes the last word in it, BAM straight onto prechorus without a breath
Oh, okay
In the same Editor, you can cut section and add a new section to create some space
After you split it, hover between two sections and there will be "+" sign to add a new one in between
Do I possess all of the rights of a cover I made under a paid subscription of a song I made with free plan?
also add [Instrumental], [Breakdown], [Drum Solo] or whatever in your lyrics in between your verses/stanzas to create music space in gens without having to edit each time
read the ToS, but if you are a subscriber you have commercial license rights. no one can really own/copyright a.i. generated content in any form as it stands. Suno "owns" it and we get to use it in our works.
this morning I encountered and unexpected surprise:
The songs cuts off before it finished processing.
My question is:
Is it possible to save those effected songs?
extend from where they got cut off
Interesting. I am novis on SUNO. Would you kindly explain what you mean and how to go about it? Thank you OBLMOUS
I have now discovered I can no download wav or mp3. something is wrong.
@native root There is a great video on Suno's Studio on their channel!
https://www.youtube.com/watch?v=qR4BefPvSiI
Start creating in Suno Studio here: https://suno.com?utm_source=Social&utm_medium=YTSF
Ready to master Suno Studio? This is the official, complete beginner's guide that will teach you everything you need to know. In this tutorial, we cover every feature: track building, editing, mixing, and how to bring your music ideas to life faster than ever....
thanks a lot !
how accurate are the MIDI exports? Like does it have CC data for pitch bends and stuff? If there's reverb on the track, will it fill it full of garbage instead of just collecting the primary note?
click the 3 dots, then remix/edit, then extend. choose the time marker on the top left where it puts yoru song your remixing. to extend from where the music cut off. This is how you can finish your song. put the lyrics where the song picks up from. Same as any extend.
Not very accurate for anything intricate. Simple piano chords and basslines it handles like a champ though
ahh thanks for the info. so really no different from other music -> midi converters then. They generally handle basic stuff pretty well
Yeah not yet. There are better tools for converting to midi. I don't think its their primary focus rn (can't speak on their behalf obv) but the rate at which they ship I'm sure we'll see solid improvements soon
the main reason I was interested is because sometimes you get a banger song, but the instruments are all mushed together and sound awful. It sounds like it won't be any help with this at all
To get it to the point where its insanely accurate would add pretty substantial processing not to mention we've gotta get to the point where the actual stems sound better when they're isolated first. Right now tracks can sound great as a whole but if you solo individual instruments the sonic quality still has a bit to be desired
I understand your frustration all too well lol
ableton is pretty good tbh, if you chop up the section then convert to midi its pretty darn accurate
melodyne even more so
but once again it comes down to the source audio
yeah exactly. when everything is layered together it's a lot more difficult to pinpoint its flaws. Though there's still some weird mixing going on. Like the acoustic guitar will be nice and warm in the verse, then you hit the chorus where the electric guitars come in and there's so much treble on it I could hear dogs barking around the neighbourhood
spot on
thats the downside to generating as full tracks
I think in the very near future though the models will get way better
eventually they'll train models on multis as opposed to quantised vectors of stereo buses
hopefully that'd allow them to separate post processing effects like reverb, so you can get a pretty clear midi export. I noticed a lot of tracks have an echo which would definitely muddy the waters
I am optimistic for the future. I'm sure they'll work it out
It will
Hey everyone. Premier user here. V5 had gotten rid of that hissing sound I was getting with 4.5+. But it's now my v5 songs have that hiss sound. I have "clean, crisp high quality vocal recording. Audiophile-mastering" in both the lyrics and style boxes... Any other suggestions?
The only suggestion I can offer that I've found has worked is to keep generating until you get a version that has what you deem to be an acceptable amount of hiss or "shimmer", then find the culprit when you export the stems and EQ it out from there
Fab FIlter or Soothe work a treat
Adobe podcast has helped a little. Trying to learn adobe audition more as well to fine tune it myself
I haven't tried the adobe stuff, but give the above methods a shot if you're able to
you should be able to get a free trial of both plugins
Will do. Thank u
Happy to help sir, enjoy your music making
Thanks, you too!
Would that be the Pro-Q4 by FabFilter? right now I am eq'ing in the default ProEQ3 by Studio One. Doing a lot of EQ'ing in my song mixes now so would be ready to level up from newb101. - some posts on Redditt are loving it, some saying even beating Soothe. sounds perfect.
The short and simple answer is "no".
What you've created is a derivative work of a song to which Suno owns the rights (the original, created on a free plan). Suno allows you to use that work as you see fit, but only on a non-commercial basis. That would also apply to the cover that you made.
If you're a subscriber, you also have ownership over the song, not just a commercial licence to use it.
Whether it's copyrightable is a whole other minefield. If all you did was write a prompt and click 'generate' for both lyrics and music then the answer is almost certainly "no".
If you wrote the lyrics yourself, then they're separately copyrightable even if the music and the song as a whole isn't. Writing the lyrics alone isn't likely to be enough to make the entire song copyrightable.
But depending on exactly how you use Suno as a tool to create the music, and the amount of direct human input and influence on the creative process, the end result may be a copyrightable work. If you factor in aspects like giving Suno the melodic arrangement via an audio upload of your own creation, editing together what's generated with a DAW with AI as a sample, mixing in non-AI generated elements, etc. then it's more likely the final work could be considered copyrightable.
The law on this is still very fresh, so it's hard to determine exactly where the line will be drawn when it begins to be tested in courts, but if you were to give Suno an audio recording of your own original work, and your own original lyrics, and use it to effectively re-perform your original song with editing and production input from you as an artist, it would be pretty surprising if the courts didn't grant copyright in the event it was challenged.
When a section is replaced, why is there no continuation? The alternates that are suggested never really fit in flowlessly.
I am feeling that pain too, when I use Edit or Studio, I am not really in control yet, hoping it is just user errror PEBCAK (problem exists between chair and keyboard)
Has faced this problem as well. For me the chorus even started while the last words of the verse was sung, so it overlaped.
We have noticed on several 4.5 version tunes that the originals have been slowed down to a tempo that was not originally recorded as such. What is going on with others, are you testing this? I'd like an answer on this surprise.
Use the Exclude Styles box and list: distortion, static, hiss, atmospheric hiss. This seems to help.
Have you previously gotten near same BPM and similar song lengths before, and now it's changed?
I haven't tested 100 generations of one upload, but as a generative AI, it would not seem implausible to expect a few to be slower and longer, other faster and shorter, and some to be a combo of both.
Whats been your perception. How many of one upload are slower? Any the same speed/length?
Seems like only the good ones that had likes. That is what I noticed. shrugs it is so irritating because it just destroys the song.
I'm more PEBBAF (... brain and fingers)
Quick question about downloads...
Is anyone else getting terrible quality downloads in wav format?
I've downloaded about seven files today and all of them are hortribly distorted and nothing like what I heard while listening in Suno.
Please definitely report this in the bug channel!
I have the pro plan but it treats me as if I am on the free plan.
It says I need to upgrade.
Any suggestions?
I also was billed for it today.
Ok one thing is for sure and I noticed that for a while now. We do not get the same spots each day everyday - quality wise. It seems that spots are not equally the same. That or the AI has some bad and good days in general. But in that case everyone would post negative outcomes at the same day which is not the case. I used the same prompts for a couple of days. Yesterday it seems I was the lucky one (wasn't for a while) and all songs turned out nice or amazing. But today "same lyrics, same prompts" outcome so far is trash one after another.
I know they got their capacity AI and sever wise but if they can't manage to offer the same quality to everyone who is paying for it is just not right. And it is clearly a luck thing if you managed to get a good spot system-wise.
Feels more like a Gacha game than making music at many times.
Can anyone help?
Have you sent in an email to billing@suno.com already?
I did.
Man this AI is just not understanding how to pronounce “logos”
sometimes you gotta misspell it, like try low-goes or something stupid like that.
I am stuck in 3.5v. When I hit 5.0v it says upgrade. Hopefully it will correct it self.
Have you tried in an incognito tab to try to rule out influence of cache and extensions?
No i haven't and to be honest I have no idea what that means.
Has someone a tip for me how I prevent every song to turn after 1 or 2 verses from what ever it was into POP, ROCK or METAL... this really is killing me in 5v...
hiii
@neat nexus thank you. I tried the incognito. It showed I was in a free plan
So I sent of an email to support. Appreciate the help.
Is is just me and the one thing I'm doing... or did someone else also got the feeling with Suno 5v the you basicly get the same song over and over again no matter what you type in style... I mean yes there is a small variation here and there but it is most of the time just replaced instruments. the Melody is always the same. The Voice never changes ...
Hi everyone,
Trying to figure out how to provide music scores in lyrics input so that we can get expected results according to desired music score.
Did anyone find how to define in lyrics, what note and scale the suno model has to use for each word in the song lyrics?
Any suggestions would be appreciated🙏
Honestly what are they doing? How can you be so incompetent? Yesterday it worked so well. It felt like a blessing for once and today all is a mess and each creation turns out like absolute trash with such weird instruments mixed in. How can they f up all the time especially if it already worked well?
ok, I've been struggling for days with v5 and extended weird intros, and I think I have found the issue with my workflow: I was including metatags for genre, key, tempo, etc before the [Intro] tag.
However, if I include a [START] tag immediately before [Intro] (and a [END] tag at the end of the lyrics) I get much better control over the intros
If you create an instrumental backing track first with deep melody instructions etc. You can then later cover it with lyrics. Just an idea if you have not already tried it.
I always get different songs if it helps.
Good to know you think it might be respecting [Start], [End] structure tags, I will try them out.
i am using bandlab ad my own masterings and it still have this branding in deezer
Even though I renewed my bill, my credit card declined because you said my bill hadn't been renewed. I renewed my bill via Android Vodafone, but it still says the bill hasn't been renewed.
you'll need to email billing@suno.com. Don't forget to include your suno @handle and transaction ID for the payment (or any other information that may help them find it)
I sent several emails, but I haven't received any response yet.
you can try and ask for follow up in #archived-ticket-help if it's been a while. just be sure not to include any sensitive information regarding your account, obviously :P
ideally you should only email them once. sending further emails may delay it even further
I just subscribed to the $8 dollar subscription last night. I paid for the year. I created some songs that I needed for a project. I started some edits and then I lost everything! I'm pissed! The prompt I keep getting is I have to subscribe to the $8 dollar to upgrade. Basically I don;t have an account although I paid for a year. On top of that I lost all of my files! What can I do?
Q. This channel is for techniques/prompt discussions. Help regarding Suno is at #support-guide and #archived-ticket-help
2. See #📚┃suno-school message
Yeah I've always found this to be an issue, you redo a section and it doesn't even flow nicely into the next unless I edit it PERFECTLY I between the transition so it does.. Plus, the music can be slightly different in production and sound
Are fab filter and soothe VST plugins for your DAW? I have a stacknof great tunes with high frequency or shimmer which really let the track down. What's the best way to remove this?
Suno Studio interface questions - How to mark begin/end of all tracks to only export that section without relying on mouse click/select which is inaccurate? And is there a way to get rid of the blank space in front of every song to start at 00:00:00? The lower right volume control that flashes white/orange, does orange mean it is reaching -0db and will distort sound?
hi!
how can i get a crminial investigation track using suno? it needs to be instrumental.
here is an example song. i'm not gonna use it, because it is copyrighted by crimesonics, but i do want a similar resault. it is instrumental.
https://www.youtube.com/watch?v=CwdllLtGoD4
the moddle i'm using is V4.5
Just in case: Did you ever in that time log off and log back in using the SSO icons, Google, Apple, Discord, MSN, Yahoo
Or you literally were editing something and it zapped and all your files went missing?
I threw some prompts down and got some vibes similar. So you could try some or a combination of these:
Slow, free expression, haunting, suspenseful electronic, pensive, operatic, aria, introspective, Electronic pop, edgy, tension, new age, arpeggio, mystery, cinematic, sparce melodic
gotcha. funny enough, on older moddles, it did a lot better with this, as all i had to type was "synth suspenseful" and it'd do it, not sure what changed in the new moddles. i'll try that and see what i get
Yep. The less you use sometimes gives better results. I did get some more pop sounding ones so I added a couple prompts to slow it down and stay minor key. Opera/operatic seemed to help.
interesting
Okay now I'm 100% confused what the F is going on... I copied the lyrics from a song into a cover... pressed create .... and the lyrics are all over the place... like in the old song as well lines or whole verses are doubled form some reason the formatting is strange... and line are some times in places where they should not... and in the last song with all female voices in a very tranquil section, I had a random dude just shouting "AH!" at me I almost got a heart attack...
aww the Suno life.. pull that slot machine handle and you too can win!
My short-film is gonna be amazing 🙂 Have the crude song the protagnist does at the beginning... then the Pop song version he gets from Suno, and finally the apocolyptic piano/cello duet Postlude version of the same song that will play at the end. 😆
I used "bassline" as a tag which worked in previous versions but in v5 it seems like it interprets that as "drumline"
Im looking for some helpful advice on how to either encourage Suno to make a referance to a bunch of songs to make an Overture or a way to encourage Suno to create several themes in one song evoking drastically different emotions also to make an overture.
Hey Everyone! 👋
I’m working with a composition created by Suno and it’s really wonderful. There’s an interlude that currently starts after Verse 2 and before the Bridge and continues till the end...it’s really beautiful and gives a “hit track” vibe.
I’d love to recreate the song as is, but also have this interlude appear after Verse 1 and before Verse 2 as well. I’m not an expert with stems or DAWs, so I’m wondering if there’s a way to do this directly on Suno.
Has anyone tried something like this or know a way to achieve it? Any tips would be super helpful! 🙏
what tips you want
Regarding the revised/refined lyrics prompts/commands to copy the interlude.
Also, is there a spearate option for Remix beside cover option?
Well, I don't know much about Suno AI, but if you are trying to create a real music experience, I can help with that. Check this out.
What does real music experience mean? I though this group is about Suno!
this songs are ai generated only
Okay. How can you help?
Tell me were you are struggling..
I don't know how to make the same interlude appear earlier, after verse 1. Currently it is after verse 2
Well.. you've told us you are not an expert with Stems and DAWs. If i am right you wanna "edit" the structure of the song.. well.. maybe via the suno editor, but i think it will be a hard process 🙂
You can try pasting lyrics after markers for sections of the song like
[Intro]
Lyrics
[Verse 1]
Lyrics
[Verse 2]
Lyrics
[Chorus]
Lyrics
Etc
hmm.. he is talking about editing a track.. 🙂
Actually yes. But I can burn more credits to try new versions, if I know the way to keep the interlude intrumentals same across. Tried Repeat but not working.
maybe you can post you lyric prompt in here? so we can see the meta-tags 😉
or the song link..
Is it okay to DM?
sure
boo no fun. i wanted to remix it.
what is a good formating for maingenre and then secondary genres just for influence? most times i put multiple as genre
[Style: Rock/Metal core; influenced by Pop melody phrasing and cinematic space textures]
https://suno.com/song/e84a5e43-c6ec-4a43-88dc-e9e76747dfdd
Owl city meets Linkin Park 😆
oki thanks
Open in editor UI.
Duplicate the interlude.
Move duplicate to before verse 1.
Check transmissions for good blends, or poor clicks or skips.
Manually adjust if needed.
Save file.
Possibly need to remaster to blend better.
you'll have to do that in an editor, Suno will not do interludes the same.
Moises.ai lets you detect a song key. 1st attempt proves that trying to put Suno into a song key fails. 🙁
So frustrating.. want version 7 of Suno already 😆
2nd attempt closer. But it put the song in Eminor.
This one worked with Key in lyrics section.
[Key: B Major]
[Tempo: 112 BPM]
[Time Signature: 4/5]
[Chord Progression: I–V–ii–IV (B–F♯–C♯m–E)]
@raw oasis sorry.. Woodhouse does not like # because they get abused by spammers to make big fonts 🙃
example
# example
@old iris
Responding to your how-to question in ticket-help:
Bringg up song/output in editor UI and click Manual BPM to set. See if that helps.
you can use tags in the lyrics to mark sections, like [Verse 1] [Pre-Chorus] [Chorus] [Bridge], etc. To play the chorus twice your best bet is to just copy and paste the entire [Chorus] section
It often understands
[chorus] [repeat]
or variations of that.
I often write lyrics for guitar or keyboard arrangement in Docs then cut and paste, and forget I have only
[chorus]
[repeat]
But it works anyway. So that's cool.
Thank you , I tray both version 🙂
Ok i put
"[Pre-Chorus]
[male and female alto vocal duet]
(Chorus words...)
[Chorus]
[male and female alto vocal duet]
(Chorus words...)"
But In pre chorus sing only male and Chorus sng only woman. Why?
I mean wht Suno make this spearate there not Duet in both Chorus?
Just know that it's never realiable to control that.
But actually its sound pretty good that if they are separeted🙂
Is it alternating male and female on each line of the chorus? Or singing in harmony to each other at the same time?
with both versions as Poltergasm writed and as you suggested, in both versions they perform separately, the first one only the man sings and the second one only the woman sings
I thought the question I responded to was about how to repeat the chorus?
https://suno.com/s/8dSzW7foPYCAdJ9S this is how I did my duet. It's mostly correct, and about as close as I've ever managed to get duets working. Feel free to try the format I used here
I also added duet to the style prompt. Not sure if that helps or not
Oh. I see. You want two chorus so they sing a chorus each, or you liked the results that they did.
Yea, cool song. I like the choice in vocal characters as well that Suno made.
Yes, I was thinking of having them sing a duet together, but this interesting solution came up instead 😄
But how exactly called this music style like this what i got there , Experimental fuson, avant-garde metal, Darsk alternative fuson (psychedlic industrial hip-hop) or ...?
btw not sure if you saw this #🔦┃server-spotlight message but you'd be a perfect fit. I always see you helping out folks
I see a helpful soul in both of you, two people who always give advice 🙂
I have one question, when I bought the pro version, there was a hint that you could make songs that were up to 8 minutes long, but now it was, that Suno didn't make a song that was 8 minutes long, he made a 4:38 long song 🙄
it can, but it's superrrr rare you'll get one that long. I've had one that was around 7:50 and it only happened once
Then Suno will be left behind for me here.
Because the longest song I've ever made - created with FL Studio+Magix Music Pro Studio+Mixcraft Studio+my tablet's music app+and my own synthesizer for effects. The end result was about 23 minutes! 🤣
Try the prompt starting with "generate 8 minutes long song...". Have never tried with 8 minute but 5 minute works sometime.
Unfortunately Suno does not work like that.
Sometimes it works, it's unlikely to get exactly 8 minute with that prompt, but 7+ should be posible.
I'm very skeptical
The 8 minutes is the maximum time based on the allotted resources (and useful for killing any ghost threads running).
Using certain features will use up that resource which results in a shorter time (see https://discord.com/channels/1069381916492562582/1422252931943501935)
Unless you are doing pure instrumental or a song that is inherently long (e.g. opera, classical), it is quite hard to reach the maximum runtime without some breaks and interludes/intermission.
For those unaware: you can always extend your piece. What I like to do for a planned extended song is to place an instrumental interlude/intermission at the end of each part so it is much easier to continue from, though that is a personal preference
https://help.suno.com/en/articles/2409601
Do you wish your songs had a different ending? You can make songs longer by using the Extend feature. Best of all, this can extend your maximum song l...
As I said it helped me sometimes, but no guarantee.
sounds like pure chance than the prompt
Befor I used that got only songs shorter then 4 minutes.
That's a good solution if Suno simply cuts the song off somewhere in the middle and ends it abruptly. However, if Suno runs through the song to cram it into the short time frame, that won't help.
I finally found this 23-minute song on an external drive, but it's a shortened version of the song, cut to 9:55 minutes. 🙂 this song/ music title I called "Wake Up"
If I wanted to have it generated as a copy in Suno, would Suno cut it down to a maximum of 6 minutes?
True but you can pad that out by using instrumental breaks or interludes (which should be mentioned in the thread I linked earlier).
There is a reason why modern songs have an instrumental portion after like two choruses
Easiest way to cut down time while retaining your structure is to speed up the BPM (ideally in an external tool so you can adjust to your liking for free).
You can always take a portion of the song you liked and upload/extend that if you want that portion specifically, otherwise I would not trust Suno to retain that structure and cut it down (not even including their upload length limit)
Do you story tell your suno style prompts? The core concept is about starting with a mini-movie in one's head. What’s the setting, main character, conflict, and emotion arc?
Example:
“A soldier flying home for Christmas after deployment, haunted by memories but hopeful for peace.”
That’s your story spine.
Structure Tags we've all seen and know what they handle but we want to put them together the way a film uses scenes such as this:
[Verse 1] – Establish setting
[Pre-Chorus] – Raise tension
[Chorus] – Emotional release
[Verse 2] – Shift or flashback
[Bridge] – Turning point or revelation
[Outro] – Resolution
Each tag should push the story forward, not just repeat feelings
Decide who is telling it and when it happens,
Decide who is telling it and when it happens.
First-person = intimate, real.
Third-person = cinematic and broad.
Past tense = reflective.
Present tense = urgent.
An example difference is akin to this here: “I rode through dust and thunder” (memory) vs. “I’m ridin’ through dust and thunder” (in-the-moment).
when moving on into the instrumentation and mood, we tell it what the World Sounds like. “Cinematic outlaw-country ballad with strings and pedal steel; male lead voice with emotional grit, building tension like a movie score.”
That’s how one writes story-tell prompts — they don’t describe a genre, they unfold a world.
#brilho
But chorus is repeated.
I'm not sure I get what you mean entirely 🙂 A chorus can be repeated...as many times as you feel like it or hit with a [Pre-Chorus]
I've thought about making music where the soloist doesn't sing, but just speaks the verse part calmly as usual. And then there's a song in the chorus where the soloist sings the chorus in a female voice, for example, but the verses are narrative and they just talk. Is that possible in Suno?
I've almost only gotten 6+ Minute long songs in 4.5 (Im a free user) the longest I've forced suno to make is 7min 56sec. But it was a combination of two songs written as lyrics that are supposed to be performed part one flows into part two with a tone change... still working on the tone change
Also trying to get a multiple tone change for an Overture style musical
How instrumental breaks can help, if it sings a content for 3 minutes just in single minute and eating some words?
Are you doing some sort of rap or -core song? Because I am usually hard pressed to make it sing/rap fast in the past for that situation to happen
If possible, can you link me to said output?
Rap, but dont want to have always "Busta Rhymes", defenetly not exactly him, but he's realy fast, so it' an example. Also sometimes the end of a verse and the chorus are overlaping.
Sorry about that
Huh. Well, the solution for this case would be to have more lyrics if you intend to have the fast-paced rap.
Hm, you can probably tell the AI to make the chorus slower too 🤔
I don't want fast-paced rap, but sometimes suno generates it. The chorus is not fast, it just starts, while the verse has not ended.
Huh I see. I rarely do rap so I do not encounter that situation, and when I try to intentionally do a fast rap, it barely does it (in the past). Wonder if there is a keyword that causes that speedup to happen
With other AIs the solution I use is to define the length of the track, so shorter length makes it faster and longer one slower.
It is much more likely to get a 5 minute song regardless, so it may appear at times, and more often, to follow your prompt though its not correlated.
Maybe you use the word "long" in your "8" or "5 minute long...." prompt, so Suno goes a little longer than it's usual lengths for whatever prompts you mostly use.
hi all !!! some problem... suno not open 🙁
Pretty certain it is a DNS issue if not cache. See https://discord.com/channels/1069381916492562582/1417459233544339537
ok. suno open after 10 min! Excuse for troubling
I'm repeating myself now, but the question was this. I've been thinking about making music where the soloist doesn't sing, but simply performs the verse parts, while calmly narrating, like fairy tales are usually read. And only in the chorus is there a song where the soloist sings the chorus, for example in a female voice. Is that possible with Suno Ai?
im wonderin how do i do vocalizing for my songs
im tryna make orchestral songs and i wanna add good vocalizin
how could I use the editor to add an instrumental intro to my song Ive been trying for like 30 minutes and it always either is completely silent or just sings the first verse
on computer on web if that helps
you could use your song as inspiration, make an instrumental version (using it as inspiration should keep it in the same key and tempo) and then use suno studio to add it onto the beginning of your song with some crossfade
unfortually I dont have access to Suno Studio , I have the Pro Plan
How can I import specific sounds and have the AI understand when and how I want it used?
Example if I want church Bell's in the opening of the song. And even perhaps I want the Bell's to be heard in first the right speaker and then switch into the left speaker?
for the panning you prolly gonna need to use studio. you can get lucky with prompting things to tell it to do what you want and see how well it does. [stereo panning, doppler effect] but even asking for sound effects can be challenging and putting [bells] or [church bell[, may or may not give you an output you like, but you can add the sound effect in studio or another audio editor/DAW
you would just uplaod the audio of the bells and place it on a new track on your timeline as usual, set the sound to come out left channel onlly. do the same for right channel on another track.
Alright. Because I hit this snag when I started to create music on Suno. So I'll need another application to make the start's then?
Studio has a feature for making sound effects, I have not had success with it myself, but it's there and imagine will be improved upon soon. but yeah, if you want a specific sound at a specific time, best way is prolly just to add it yourself in Studio or your DAW of choice
otherwise you can use the text tags in your lyrics and experiment and see what Suno gives you.
could create an instrumental with bells and then clip what you like. (if you can get a bell instrumental) seen people having a hard time in here with bells specifically
Then I'll have to manually create my starting sound and Effect before then cutting and pasting it into the tune I wanted to make then.
Text tags in the lyrics? I need to look this up! I understood you as I can use a tag in the lyrics to perhaps force a sound to trigger?
in your lyrics section if you wanted an instrumental intro with bells, put that in meta tags where you want it. (in thsi case at the start)
[Intro- Instrumental, Church Bells]
anything within [brackets] is treated like instructions for the ai. it reads that and will try to make that happen where it is placed in the lyrics prompt of the song
no command codes needed. uses natural language. but some words have more weight than others. general overview of use of [meta tags] reccommended if enw to it
Alright because when I tried that in the lyrics. What happened was that the AI just sang them for me haha! Sounded super strange
it will sing if in paranthesis (backing vocals) [brackets should not be sang]
That's super helpful honestly! I'll try it out!
Is their a way I can check if the AI has a sound source on hand already?
Church bells being one thing.
But I wanted to use a sonar ping in the intro also
i have not searched. i only know it has been able to do stuff that I tried. [laughter] [sirens] [deep space soundscape]
Thank you!
going to try for Sonar pings
The exact prompt started with: "Generate a 5-minute Rap Hip-Hop track".
Sonar Ping is tricky.. but I dig the experimental sounds - some clips in there to cut and experiment further.
https://suno.com/s/719XQLkA34hIufEA
made me wanna give it a go... i got some sound effects.. but not a sonar pulse/ping or [Subamarine ssonar sound effects]... but if ya need sound of some bicycle spokes.... i got you. 🙂
https://suno.com/s/mfL5T6lwRwyEnWlT
Since the very use of different genre prompts can change output length, then, if it works for you, go for it.
Thanks for always trying to help others in here as well. It's appreciated. We can't always know what will work in every case but any suggestions are usually based on experience, so all we can do it try a few and see what works best.
Tried it with some other genre as well.
https://suno.com/s/tsD1BdL3xsut9q3v
Used Studio FX with (Sonar Ping) better results. Jut noticed I had it on v4.5-all earlier, but whatever.
Now that is a good setup! Nice
If I request a time length, but give it just4 lyrical lines, it's likely to be only a minute or two, as Suno seems to keep tempos relative to the genre. So four lines of lyrics and a 90bpm rock song may be result in about 1:30 at the most, along with one or two 4 to5 minutes and a surprising one nearly 8 minutes. Suno likes to vary things with short lyrics.
If we say " make a ___ minute song" and it works most times, then in theory, if nothing else changed, we could change the time and get a variety of different tempos, or else a lot of filler instruments or lyrics repetition.
If that's true and worked, that would be a handy phrase to use.
Take away all but a couple of your lyrics and see if the "make a 5 minute rap" still gets a 5 minute rap.
Anyone have a high-soprano female persona yet?
challenging to get.
Getting tastes of it.. but I want MORE... "Sing for me.. my Angel of music!! Sing!!"
https://suno.com/s/EEVpHbGdzksTOJFy
Hello everyone
doe anyone know hwo to put your finally gfinished songs on spotify or if there is even a way too? Or if we should?
I have a problem with the renew of the subscription. When I click on "subscribe", the redirect doesnt work. I changed browser too, deleted cache etc, but anything happens. Thanks!
it doesnt work neither on iPhone Safari
looks like my account has been blacklisted or something like that lol
What happened to the first subscription then?
Gues you need to have spotify account first.
"Upgrade now and this song will upgrade too"
Does that mean I can have commercial license if I upgrade?
Since free model is now 4.5?
The link to the FAQs says otherwise. Is it not yet updated? Or I am misunderstanding the sentence I qouted?
from the way i read it, thats how i understand it. model isn't importnt. "4.5-all" is just the new model for all free-mium users. upgraded from previous model tier.
So do I get commercial rights if I upgrade?
to that song you made on free-mium/ i'm not 100% just agreeing with you that is how that screenshot reads to me. it probably only for that song, not any previous gens. just the song you made when you clicked that link. i presume.
"this song will be upgraded too"
I wrote a song with a melody in mind but I tried Suno and it did an amazing job interpreting my song but I am on free mode rn.
That is what bothers me, it feels like a new thing they implemented but the answers on FAQs haven't been updated or anything
it looks like a promo/incentive to get you onboard. If you created a song and you really liked it, gives you chance to upgrade account to get license for it. Wasn't an option before. Nice to have the option.
I will upgrade but just not now hehe
is there any way to get covers in v5 to be as good as they were in previous models?
or is it a known regression
Downloading Tempo Locked Stems or only Stems (no midi) doesn’t work. It just downloading the Drums file. Stems download only works when using the Wav+Midi or Midi download.
This is a big issue that the tech guys will fix ASAP… You can’t even download Stems in Mp3..
Can anyone recommend which prompt style to use for v5? Either a list of terms (dark, ambient, slow, drums) vs a descriptive sentence (slow and dark ambient with drums). I know theres a built-in tool to generate a descriptive sentence prompt but I don't have much luck with it.
Can Suno add like sound effects in a song, like car sounds etc?
i recommend talking with a chatbot about what sounds you are trying to create. asking for musical terminology etc.
it can, but it's experimental. can use meta tags [car alarm], but it's not a guarantee. can use Studio to create effects, but it's still Beta and same with being unreliable. better to just instert sounds you want in the song with Studio, or other DAW
Hope the Tech guys fix the STEMS download.... Just the WAV+Midi download working
@torn spoke showed me this. you can use this to split your stems for free until Suno working again. Just have to upload your stems.
https://huggingface.co/spaces/TheStinger/UVR5_UI
The WAV+Midi working as I mention. The thing is that I want to use the new Tempo-Locked version...
I posted in bugs after I saw your post, and tested trying to download mp3 files myself, and it wasn't working for me either. Hopefully fixed soon, like you said. Tempo-locked version sounds like could be very useful for people. New features, new bugs.
Is there a list of commands I can use in lyrics?
Is there a way to make animal sounds?
hey folks general question perhaps someone can help out. I am attempting to salvage an old 3.5 track that i fell in love with and i am struggling immensely on refining it to be remastered without losing quality. about 200 gens in over the last couple weeks putting it down and coming back to it now and again. What is the best slider combination percentage, and or prompt instruction to retain say 75-90% of the original structure and sound without losing it or the quality. if it's not possible i will have to keep chancing cover after cover. but no luck thus far, Thanks!
Nope. You don't get commercial rights on any songs made on the free plan, even if you get a subscription later.
Who told you that
It's always been that way. Read the TOS.
in the screenshot they showed it looked like they could upgrade and it would give them the song for upgrading.
how can I get an instrumental on v5 to the longest length without having to extend them?
also on the other end of the scope how can I make looping short tracks
Sounds like you're Pro tier now? Have you downloaded and uploaded, whilst attempting to Cover as close possible. If not try once, with audio slider about 98%. You can leave the others alone or set them 1%.
Yes Pro member, I shall try it. it's super tricky i am finding. what it is it was an upload following a bunch of audacity splicing multiple liked versions together. i think it was 3 covers that i liked parts of each and spliced it to the point I really liked it and re-uploaded it. am i better to cover that one? or find the very original track to attempt the same. extensions? there was a lot of outside work on the old one so I am hoping to allow that to carry over and not require it to be started from scratch
Interestingly, we can say the best quality audio, the better cover. But there is also the idea that Suno can extrapolate and interpolate extremely well, so a poor recorded or glitchy, distorted audio could come out sounding better.
Try the one that the arrangement and harmonics sounds best, regardless of which has the better fidelity .
fair, that heavily edited one at one point was nearly released and went through a pretty thorough remaster system so it is clean, but just a lesser model. i'll try that low percentage now and near maxed setting now.
then newer models came out and i've held onto it since. then with the 4.0 shimmer issue i still held off as those covers still had traces of it as well. 4.5 i've began dabbling in it again and with a successful other 3.5 rebuild to 5.0 it got me hopeful again.
This is the prompt I used to generate my song the other day, but I use the same format every time. Keep in mind that you only get 1k characters. Also if your song has lyrics, the section headers should look like [Verse 1 - Vocal instruction; Instrumental instruction]…for example: [Verse 1 - low sultry vocal, dark trap beat building, subtle glitch fx]
Genre: Dark K-Pop Trapcore
Mood: Seductive, tense, glitchy, vulnerable-but-dangerous
Tempo: ~145 BPM
Dark K-pop song about someone mentally undressing and exposing all hidden secrets. Seductive female vocal with rising intensity. Whispered ASMR moments in intro and bridge. Powerful belted chorus with layered harmonies. Heavy distorted 808 bass, glitch percussion, metallic hits, and eerie pads. Digital stutter FX on post-chorus. Cinematic and cyberpunk mood.
Title: “SECRETS (Undress My Mind)”
Production Guide
• Vocals: Whisper → breathy → powerful chorus, some robotic chopped repeats
• Beat: Dark trap, sharp hats, glitch fills, aggressive drops
• Bass: Distorted 808s, sidechain pump, sub-heavy
• Synths: Metallic stabs, airy pads, cyber FX, filter sweeps
• Mix: Wide stereo, vocal forward, glitch elements cutting through mids
Hope that helps.
Thanks!
Are you putting all of that (including production guide) into the style window? Your prompt looks a bit similar to what I've tried before but never had much luck. Although its hard to tell since the generation is randomized.
I only do instrumental music so I never have to worry about lyrics or the text limits
Yes all of that goes into style. It usually gives me similar to what I was envisioning for the song. And most of my songs have a darker mood.
Sounds good, I'll give it a try. I didn't realise that you could list of different components of the song, I thought everything had to be in a large paragraph separated by commas
Use the Custom option instead of Simple.
yea I use custom, but I assumed anything out of the "item_1, item_2, etc" format would just break it lol
Is it possible to have certain words in lyrics been said with emphaisis?
you could try CAPS
No need to use sliders — just upload the song in WAV format or play it from Suno, remove the style tag (it must say “no style”), and generate. With Module 5 you’ll get the same structure of the song, but produced even better.
Amazing, I'll try this! Thank you.
generate - you mean Cover?
What you think about this latest news about buyout of competitors?
I hope Suno won't give after
i have Issue with Extending Music Using Suno
someone knows a prompt to make a verse sound "robotic"
like an echoed vocal, a literal robot singing
i tried robotic and echoed and nothing happend
i will try
just with robotic voice or robotic effect tag. Try add experimental to the style, or maybe electro-experiment. Also, try using another model like v4
upload sample, that's the easiest
i want just a verse
but i got something similar
Anyone else have trouble adding new vocals or new speech to a cover? I put [spoken words by an inhuman, guttural demon: I am at war with the world and every living soul in it, I will kill them all.] I wasted 100 credits so far, nothing is happening. https://suno.com/s/L6afHun3EHADiPkz
I'm tired of paying for this.
Trying to do a Pan Flute, Piccolo, Soprano Flute trio... I think I might be pushing Suno too far.. probably is not capable of such specific instruments and I would have better luck programming into a DAW (timewise...)
It doesn't make a difference, since the way Suno "reads" both (and may ignore elements of both) is functionally identical.
Personally, if I know exactly what I want I'll usually set it out structurally, similar to Christin.aiko's example. If I want Suno to play around more freely and see what comes of it, I'll do a shorter narrative prompt with some key info only.
The simple route is to Remaster with 'subtle' variations, although I find v5 does still like to exert it's creative input in remasters.
The alternative route is to Cover with audio influence set high (75%+) and weirdness low (<20%). Style prompting you may need to play around with, depending on whether your prompt accurately describes what Suno actually did (if yes, you can set it high - if no, and be honest, set it low). You can also reinforce the original sound by making a Persona of the 3.5 track and using that along with the Cover to further reinforce that you want the audio to sound similar.
It doesn't guarantee you'll get what you're after, but it does tend to tilt the odds in your favour.
There is no way to reliably ensure Suno switches correctly between specific vocalists, but if you search for "duet" or "male and female" in here, you'll find a number of suggestions for how you can try to make it more likely, or work around Suno's limitations using multiple Extends or taking to external editors.
As an individual, you can't directly upload to Spotify - you need to go through a distributor. Spotify provides a list of it's distributor partners that you can browse through, but as an independent artist most will charge you some kind of fee for managing and maintaining the distribution.
In terms of "if we should", as a general rule you should never upload something to Spotify which you know somebody else has ownership and/or copyright over unless you have explicit permission from them (ie, don't upload songs created for free on Suno, or anything which is based in any way on a pre-existing copyrighted song).
They [Flipx] are saying Cover in 5.0 but leave the sliders at default. I say try audio slider at near 100%. Vary it between 25% and 100% of preferred to learn what works.
It's not always about one technique is better. Suno still does what it wants. You pitch. It hits.
Our style of pitch does not necessary make the Suno's swing a guaranteed home run. Even our best best fast ball (what's work for others or us before) , Suno may bunt.
Don't forget, Suno doesn't absorb the WAV file in digital notes like MIDI. It processes the waveform. So you upload whichever gives you the sound you like. If mp3 version sounds better than the WAV, use the Mp3 or mp4.
Whatever you hear in your ears, Suno hears better, all the ghosts in the machine that don't always show in the more straightforward, no frills, uncompressed WAV file.
Some home runs are hit off of a curve ball.
Explore. Have fun.
But the old suno model was much beter, when reusing audio, no meter if uploaded, or some existing one.
Yes, cover option
Hey guys. I've just cancelled my Pro subscription. My nose is a bit bent out of shape thb. I didnt use Suno fr a couple of months and expected my credits to have stacked up, Heck, I paid for 'em, only to find the credits simply seem to cap at the months subscription limit. Is this a bug or intentional?
I have songs published that are not showing up on my profile even though I have published them. I have the pro version.
credits do not carry over
It's not a bug. You have to use your credit within the month. It doesn't carry over next month.
only credits you purchase as extras don't expire (but you still need a sub to use them)
Righty-Oh
it does warn you in tiny text
👆
Hi what must i do, that V5 make songs longer as 1:00~2:20min?
I very rarely see them that short. You may be exaggerating.
or you're using a fast song style with not many lyrics
Or posibly just instrumental.
the songs I posted above in the screenshot are instrumentals
@robinclarkblackboxdigital where is no style option. Do you mean slider at 0?
But I also had the same problem only with instrumentals, but it was with 3.5.
I have try it out with long Songtext, short Songtext same. i cut the Style of Music on 2 Styles and same, then i try it out without Songtext only Instrumental. 2:20 again its not funny to waste Credits
As mentioned befor, what helped me is a prompt like "generate a 3 min track".
Hi there! Can I still download the wav files till the the of the plan if I quit my plan now ?
Why Suno Ai put always In to intro something hums. This is so disturbing just. Iput in Advenced Options : male voice, male vocal, arabic, africa, , AI inspirations, hums in inrto and vocals.( arabic because that if I try make celtic or nordic style, then its make like its sonds arabic not celtic or nordic).But still every time it put there in [Intro Instrumental] the hums or some another vocal. and additional i put in the style there "AI-inspirations, or creations are strictly and categorically prohibited." and because that i have only 65 cretits now 🙁 I've tried genrate the same song 16 times and it still starts humming.
Anyone having an issue where you can't replace a vocal section in suno studio? When I try to regen a second verse, it generates no vocal at all. Using v5.
Curious if there's a trick??
Yes you should still be able to.
100% sure ?
you retain the full functionality of your plan until it expires, even if you cancel
It does happen from time to time.
Maybe you should try to find another way to regen that verse... maybe use Extend instead.
Then once you're happy with that new gen - either go with that new gen or bring it in Studio and use that section only.
Suno Ai could do the opposite, that if the humming is not specifically noted in the introduction, it will not work. As if an AI algorithm had been written in there that the story starts with a humming. That's the impression I get, sorry.
There could be a similar separate box where you can write down what the AI can do. For example, there is an "Advanced options" where you can write down what the AI is prohibited from doing (which is not recommended).I think it would be much easier for users to create music this way, with the help of AI
@half dragon but it's happening every time I try. I've tried different songs, different sections, different settings. Seems so weird since it's literally one of the big selling points of suno studio. Trying to figure out if I'm doing something wrong?
And can you "extend" in suno studio? Isn't regenerate just extending?
btw idk if its right channel but anyone has idea why my suno suddenly when everything is in english the versions are changing randomly language and stays in those japanese/chinese sticks?
I can still tolerate Suno's AI 🙂 , but the AIs that I wouldn't want to use, but have to, are Google or similar translators, which completely turn your sentences upside down (arbitrarily replace words, without knowing the exact linguistic structure and properties, etc.). Google used to be many times better and more accurate than the current Google translator with AIs. I've already tried to translate this text from Estonian to English 7 times and somehow I managed to do it now, I think.🙄
Yes. Technically, it can do all that and when it did, it's brilliant - but it's not perfect and will improve overtime for sure.
I'm sure you did everything correctly - select what type of gen (song, instrument, or vocal...), then style prompt, then lyric - like any other generating tool, like Extend, Cover..etc...
Both Editor and Studio can extend but I found the Extend tool is more reliable.
Interesting. Randomly like sometimes it's in English? and other times, it's in Japanese?
Have you tried to clear the cache to see if it helps?
If nothing works, report in #🪲┃bugs-web
when i refresh the page and quickly open the version tab it randomly switches between languages and stays in german or japanese idk why xD
Oh easy then. I do not base things on copyritten material and I don't use the free version of suno. I used it three days a year ago then subbed and haven't gone back lol. XD Thank you for answering!
Hey i got a question for either a tech or those whom have a subscription is there a way to use your own voice for the vocals
Use the exclude box and tell it what don't want. Also prompt in the Intro tag. [Instrumental Intro][No voices][Instrumental Only]
Why is it not recommended to use the Exclude box? No one has ever said that before. You just need to know how to use it.
Hey, I worked on a song in the studio and there it's perfect. When I'm transferring it to Suno it sounds terrible. Distorted. Why ?
are the levels clipping? (make it wide enough to see the master fader and see if it's peaking above the 0db mark, it's orange/red it is)
Where can I see it ? The master fader ?
sec. making a screenshot
this is bad 🙂
it's ok to hit orange occasionally... but those 2 little tic-marks that show peak, that's what you don't want
How can I fix it
record yourself singing, use turn the upload into a persona. Persona tries to use the vocals. So if it's you singing, it will try to sound like you.
go thru and turn your individual tracks down. as a general rule I lower them all 5-6 db and then mix accordingly.
Ok thanks
Hope it helps 🙂
🙏
I muted the voices I recorded and kept only the playback and the master fader is almost white all the time and still it's distorted
hmm.. that's the only thing that came to mind, haven't seen this. if you want to dm me I'll see if I can help you find the problem
I find I have to go into advanced options and type in, humming, moaning, swooping, soaring, anything you don't want to hear in your song. Include in styles; plaintive tone, raw, authentic, detached. It has worked for me hope it helps
i did that, but still same.
I really dislike the humming thing myself, for a while I was getting a really awful cooing sound... sorry I couldnt help
I just read up on the how to prompt a song longer and I have tried a few things I cant seem to get the extension to actually make longer songs either was there any resolution from earlier on that question
suno is like ignoring all my lyrics and all my highlighted section and just doing a super short clip that cuts of mid word no matter how many times i redo the edit section. I tried with the new launcher and the legacy i'm getting annoyed at the wasted credits. Anyone else having this issue?
u checked you lil box? (sounds like you're using editor)
lots of lyrics and genres that involve longer pieces of music
i say lots of lyrics, but usually 5k isn't enough for me, with minimal prompt tags in lyrics.
i had it ticked, i'll try again i had the page pulled up for weeks so i closed browser and reopened hopefully that will help too.
thank oblmous
thank you...
extensions dont really necessarly make a song longer or can they
dude like out of nowhere suno makes a 5 minute song idk
thank you sunojesus
extensions do exactly that. make the song longer, extending it.
if you're using cover to create songs, you have a lower cap than the 8min cap with a normal gen(subscribers, i dunno cap on freemium)
sunoesus
oblmous my extensions the last hour when I opt for Full song it just gives me a shorter song
idk whats up
sounds odd. only thing pops to mind is that extensions have parts. maybe you thought you were getting a full song but didn't follow the extension path you were on when generating. hope that makes sense.
original generation > extend (paart 1) > extend off that part 1 extension to create a part 2. when you get done extending you sometimes get lost with parts if you were jumping around with the extensions
oh shoot thats what happened
so wait lol
so then which one do I need to get the full song from
dude why is this so confusing lol
i think im getting old
you can extend for a long time... gpt was searching internet for me about other info and came back saying something about the time caps, and not sure there is one. quality degrades after a while tho typically
you want to get full song from the final part
like the final part
extended part 2 then get full song then it should be og plus the extension
the last extension you made will create Full song starign from originial gen, through all the parts you extended (these are all in you workspace). But only the path it followed to get to that extension.
whats crazy is about 5 months ago I amdea bunch of music extneded all of it worked great no issue somehow between now and then i got dumb
where you selected the timestamps from each time you extended. it's pretty straight forward, but can get turned around easily with multiple extensions aand multiple parts
okay so then help me with this sir if you would please and I thank you
what does setting the timestamp do in the extension process
i dont get it lol
you're choosing the starting point of the extension.
and its just one stamp right one time
(choose within music if possible, rather than a point within singing, or speaking, for better results)
you can type, or drag. this is where the extend will begin from the track you are extending. it stays put while you hit create, but you can move it.
sounds like maybe you're thinking to linear. start thinking branches, not lines.
everytime you create a new extenstion. it's a new branch. branch from original song, and then new branches off the extenstions you make.
Hello,
How can I improve the Rally House prompt?
Does anyone have any information?
is your current prompt just, "Rally House"?
Yes, I really like that style (in an interesting way), I wonder how I can make examples in suno.
i had to figure out what it was a bit, so it merges 90s house, PS1 racing soundtracks, phonk vocal loops all wrapped in 80s/90s rally aesthetics?
So chatgpt starts suggesting something like:
Rally House — high-octane 90s house with PS1 racing-game soundtrack flair and phonk vocal loops. 125 BPM. Four-on-the-floor 909 kick, crunchy breakbeats, FM bass, chorus-washed synths, and chopped vocal samples. VHS-style saturation, car-engine ambience, and neon-lit rally-track atmosphere. Feels like Gran Turismo meets Phonk in a warehouse rave.
Edit as you desire. Interpolate. Extrapolate. Good luck with the ps1 racing soudns. that is something you may want to upload a sample of and create from it a bit to blend/add.
Would try [Car engine revving] or anything to like that to inside the lyrics. Inside the [brackets] in your lyrics prompt to see if can get it to do some [Tire Burnout] sound effects with meta tag triggers. But there's not telling, as far as I know, what sounds could be trigger. Lyrics wiith v5 effect soundscape now too. So lyrics that go with the same ideas may help with trying to create the car soudns sspecifically within Suno
Thanks, I'm Trying
or try to "reverse engineer" the style wiith an artist with chatbots. Is there a popular Rally House artist?
A glitchy electro-pop track with choppy, syncopated drum programming and angular, processed synth stabs forms the backbone. Verses are minimal with tight, stereo-panned textures, while choruses burst with bright, layered melodies, jittery samples, and stuttering vocal FX.
This is something in the style of DXCD77... for example
hmm
Fuse them all together with parts you like, see how Suno interprets them
I Trying, thank you very match
Good luck. Hope get some good tunes
I think I may have found a quick workaround for anyone that wants to change the cadence of their songs. A lot of people I see say the song is going to quick. Try using the Adjsut Speed, small adjustment. Even if this doesn't give you the timing you're looking for, using the Cover feature on a slowed down version of a song you liked, but was too quick (or vice versa) looks promising with what I'm experimenting with.
nice
okay everything i a generating it turning into 2 second clips this is so infuriating!
I just want to be able to do the end of this one song!
I only saw now that together with the stems I can download also the midi tracks. How good are the midi tracks actually ? Anybody who used this feature ?
You may need to backup further into the song, or use other features like Replace instead of Extend. Or even Cover (adding an extra verse at the end).
I was trying to redo the last 30 seconds of the song. How did you know? hahah I used replace not extend thats what i was doing I was trying to redo the ending cause the end of the song was bad. But the rest was good. I was in fact not using extend at all.
is it possible to make a song fully acapella, or remove the instrumentals and just keep the voice lyrics?
Anyone got a nice guide/crack the code to prompts forr studio?
Such that it actually generates a single instrument etc. hahaha
It is possible, but it is luck. I managed to make some but not only with one voice, but with some other vocalists
I clipped this from the end of my pop song version. I created my Postlude version with it
https://suno.com/song/ce74f6a8-eb17-44dd-94fe-280f5e88b1ed
"piano solo; no other instruments"
https://suno.com/song/965ae1d6-de69-45bd-8801-8044cc80cb77
later remixed it as a Piano/Cello duet and released it.
I guess v5 is nerfed 😄
That is a common theme I hear.
When I had this issue with V4 they suggested starting the extend at least 6-10 seconds before the track cuts off. Preferably in a break between structure tags (example between [Chorus] [Outro])
Or add
More
Lyrics works
For me
Can anyone tell me how to make the chorus more powerful? I just want to change a single final chorus to make it stronger. more impact
Try to use [final chorus] instead of just [chorus].
Subject: Account Flag Desync – Pro Plan Showing "Upgrade" (v5.0)
Hi Suno team,
My Suno app is updated to version 5.0 and shows my Pro plan (next payment Dec 7, 4900 credits),
but the interface still shows an “Upgrade” button and gives “Internal error” when I try to generate music.
I’ve reinstalled, rebooted, and cleared everything. Please manually refresh my account
permissions on your server.
sounds like a support thing #support-guide
@wise kindle thank you. I did and they never got back to me.
you can try and ask for a follow up in #archived-ticket-help if it's been a while
I@poltergasm.official I did it last Thursday. I sent it again just a bit ago snd got the same standard reply. I thought maybe coming here someone from the Suno team would see it. Thank you
ah I see. sorry, but that's about as much as I can tell you. hopefully someone from the team will see it and push it through, but there's no guarantees
@wise kindle . Thank you so much for responding and for your help. I really do appreciate your help and kindness.
Anyway to reproduce the sound of "talking drum"
My pleasure. I really hope you get your problem sorted soon
btw Joe, I don't see a message from you in #archived-ticket-help. Don't forget to try there if you haven't already - it can be useful if you need a follow up to a support ticket as a team member may see it
Wish suno had more instruments support like the African talking drum
Have you tried dùndún? idk if it'll know what that is, but it's worth a shot
@wise kindle do I include me email address there as well?
No, I wouldn't share personal information. Your suno profile name is probably enough and a brief description of the problem
I tried that but it gives paino plays on v5 and v4.5+
@wise kindle thank you. I love the name BTW.
mmm does it sound anything like a djembe? that might be a little more common than a dundun
@wise kindle Ok I sent it there. At least I hope i did.
yep I see it in there now. good luck!
@wise kindle Thanks buddy.
Sounds similar but instead of using hands you play with a stick which gives a different distinct sound, thanks for the advice though, I will experiment to see if I can get what I'm looking for but moving forward I'm curious as to if more instruments will be added to future versions and can we add suggestions so eventually all instruments could be supported
Good day! Could you please tell me how to clearly separate female and male vocals in a song using Suno? Specifically, I'd like to assign different genders to certain phrases in a verse within a single line.
Ah ye I see what you mean. Yeah unfortunately instruments like the dundun are pretty niche, so Suno likely won't have a lot (or any) support for it. Sometimes you just need to try and pick the next closest thing and roll with that. I hope you find the sound you're looking for
You can try and coerce it by using tags in the lyrics, like [Male]la la la [Female]fiddle-dee-dee. It's not guaranteed to work the way you want it, though.
I've already tried, but it doesn't work...
[Male Vocalist] Hot lips meet mine—my dead heart suddenly beats!
[Female Vocalist] "I won't leave now..." [Male Vocalist]—her wow the silence greets.
But thanks, I'll keep trying!
It may also help to let it know you want a duet by adding "duet" in the style prompt. Again, I'm not too sure about this, but it couldn't hurt
and Try Male Singer/Female Singer instead of vocalist, or just Male/Female
OK! Thanks again!
ok this has been killing me. So I have re-rolled this song like 20 times. Most of them are great and followedx my prompts great. Problem i am having is each song the male vocals are upper register in pitch. How can I prompt to have it lower the singers register?
you could try the keyword "baritone" when describing the vocals/singer
suno is fairly good at recognizing words like that
hello all, is there a way to upgrade personas from a previous version to 5? i have premier but my brain isnt functioning this morning lol
i remaster the original song i used to create the persona, but i dont have an option to create persona from remastered version now
You can fade just a bit and create new persona I think
i'll give that a shot
somebody help me?????
My cover is 5 minutes long, why does Suno generate less than 2 minutes of music? Is it related to the cover being longer than 5 minutes?
I am so fed up with this Ai not doing what you tell it to do.
Yeah, these hands ain't clean,
Been dryin' a thousand tears.
Instead of singing those lines fixing it in Studio it only sings the 2nd line all the time leaving the first one out. I already generated 20 times and it is the same all over again. I freakin hate this. There is more than enough space to fit in those 2 lines because it originally was their dedicated space but because freakin Suno decided to sing some dumb gibberish which never will be fixed as it seems I am forced to fix it in studio which also works really GREAT! 🤬
Help! I encountered this error while creating a hook: Exception during hook creation: Your account requires support assistance.
If you don't care about the key, adjust speed down, say 94%, then click "keep tempo" and adjust speed back about 1.06%.
i'll give it a whirl
curious, does it sing, " Yeah, Been dryin' a thousand tears."??
I ask because if it seemed to sing the "Yeah" and then skipped your "these hands ain't clean" maybe it's due to the commas? Maybe this is an extension you're doing and timing related, but in writing what you see between the commas could be left out and the sentence still read. you know what i mean? i wonder if it was taking the option to skip lol
I even erased the "yeah," part and still the same. The only thing that worked (more or less) was as I generated them one after another which means first "these hands ain't clean" alone - without the "yeah," (just to be on the safe side and then again the 2nd line "Been dryin' a thousand tears." but in both cases only one came out fine all the others gibberish and terrible nonsense.
what gen method? extend, replace section, og create
Replace in Studio
we definitely need a build-in De-Esser Plugin in Suno Studio 😄
at least your not having issues with the heteronym "tears"? 🙂 I having my own gibberissh issue with a song lyrics. Trying to figure it out as well. I suspected mine was because lyrics say things like "language irrelevant" but every gen the lyrics in the same part went like Novocain mouth.
Someone please help me, I have a 5-minute cover, and Suno only generates the song at 2:28. Has anyone else had a similar problem today?
Mine "sang" such weird stuff which was neither part of the lyrics nor the prompt something like "This is not a modern clean recording" which makes no sense. No other generation sang those lines. Just freakin weird.
I am still receiving a persistent "Create failed due to an internal error" on my main Suno account:
I previously contacted you about a similar issue that occurred after using a VPN, which I resolved by clearing my browser data. However, that fix no longer works.
Crucial Proof (Please check my account status):
-I created a new Suno account using a different email.
-The new account works perfectly on the exact same device and internet connection.
-This confirms the "Internal Error" is specific to my main account's data on your server.
Please check and manually reset/repair my main account's profile on your backend.
Thank you,
is there anyone to solve my problem? I got an e-maik to suno but They didn't answer
suno support via email needs up to 14 days for an answer... be patient 🙂 and no... i never had that problem
There is nothing we can do for you in this channel. Please post at one of the bugs channel, and you can contact the support team (see #support-guide but response time is in weeks due to massive backlogs from problems in the past month).
Is that one of the Suno app? Make sure you have the latest version. Sometimes, something in the app may be broken and you would need to use the website to do your work until it is fixed...
ok, ı see. I'll make a contact with them again, thnx
If you already emailed support, please don't email them again. It will not make your response faster and may actually make it take longer to get a response
Hey anyone is making rap music and can help me?
i think we can answer that if know the question 😉
im having hard Times with persona, i want to get the exact same voice for a rap album but its not exactly the same, and it also copy the style of the beat, any help?
Hey, all. Q: If a track generation is all-but perfect, but skips a beat a couple of times, how do I 'repair' the skipped beats without losing the rest?
(So frustrating. It's been great all of this session, then just when it nails it, it also has a couple of skipped beats. 🤬)
sry, but i don't work with personas
...
separate the vocal stem from the music, cut out any artifacts of music, then turn that into a persona. It's not a cloner, it will always vary.
same as anything. if the option isnt there it is a bug. open the stem in editor, save as new song, then you can turn that new file (copy) into persona
So you can never get the same voice on multiple songs ?
On phone theres no option to do so
voice depends often on the style... if you use the same style prompts the vocal model is near up to 90% the same 😉
soudns like trying "Repalce seection" in ediitor mayu be worth exploring. depending on length, would want at least liek 12-15 seconds prolly to work with.
of course "reggae vocals" are different then "pop-punk vocals"
Thanks, I'll give it a try.
I would think of it all as more Infuence, not use this. That would get weird like Robin says changing vocal styles, or pitches would create problems with text-music stuff
opening my phone suno app to double check
Having the same voice is the most important thing when you try to make an ai music artist.... Its deceiving
well id unno if youre on appp, but open your phoens web browsser and go to Suno form there
the problem is, that nearly every "ai artist" would sound the same 😉
there are other tools you can use in tandem if you are wanting to create a virtual ai artists. that is a whole different bag than using ai to just create music. Persona is pretty good for what it does. ElevenLabs and other ai voice programs would have to be used to have a consistent voice across the board without a human involved in actual singing
Hi guys, can you help me how to make a song that are similar to a particular band or artist?
what do you mean with "similar" ? Style or vocals?
Not at all its the opposite, as soon as you get a very unique voice you can use it on all your songs, theres no coherence or consistency, no one would believe if the voice inst really the same
Everything, i want make a song like greenday, basket case but suno create like simple plan
that's true... but it's still the same voice model overall 🙂
forget it 🙂 suno is not an ai to "clone" artist vocal sounds... and the style of music can be created with the prompts...
I dont care about vocal just music
if you wanna do some greenday stuff i think the prompt "pop-punk" or "2000s skate-punk" will be the best..
Thank you, i will try
if want green day basket case, but getting simple plan, then your in the ballpark of punk, but you need to detail and articulate what about basket case you want to emulate. the bouncing bass line descending down the scale?
@scarlet tendon a prompt could be: High-energy pop-punk instrumental with fast tempo, crunchy overdriven guitars, tight power chords, and punchy drums. Driving bass lines and catchy melodic riffs full of youthful energy and rebellious attitude. Clean yet powerful production with a slightly raw, garage-band edge. Evokes early 2000s punk rock and alternative rock vibes, mixing emotional intensity with upbeat, anthemic energy.
I see
Wow cool, thank you
of course you can add something like "intense powerful pop-punk vocals"
2000's punk 🧐 early 90's!!!! 🙃
yaaaa..you are right 🙂
time goes bye sooo fast 😉
@scarlet tendon : let me know if the prompt is working or not 😉
It's good bro, better than mine. I will use this. I like the result
great 🙂 and it depends on the model as well... if it works with v5 it can be an absolute mismatch with other models 😉
I’m creating songs with Suno, and while the track turned out amazing, some of the pronunciations sound strange.
For example, I want it to sing “Yes,” but it ends up pronouncing it like “Ye Yes,” which sounds obviously AI-glitched.
I’ve tried remastering multiple times, but the same part keeps coming out wrong.
Is there any recommended way to fix this?
what model?
V5
hmm...
This is a pretty common issue — I’m sure others have experienced it too.
First of all, thank you so much for taking the time to reply. That alone already means a lot to me.
well.. i personally don't have this problem till now... maybe try to cover the track... remaster often use the same vocals while cover re-creates it
Thank you, Robin. I’ll keep trying and won’t give up.
good luck 😉
Can I send you a friend request?
try it 😄 😄 😄
si i figured out that the legacy editor works way better for word replacement. So now that i have all my edits in how do i save it as a new song in the legacy editor?
when youre using legacy editor, when you click "Apply Replacement" it replaces that section within the track and your timeline is replaced with the song with the new edits. These are saved automatically and found in your Workplace
got ya thanks
hey folks, how are you? How do I change the lyrics of an existing vocals, I mean to keep the same style and singers, but change the lyrics. There are some typos 🙁
Anyone else having these stupid hiccups and short stutters in Vocals especially within the last days with V5 it ruins really great songs because of this crap.
could try using the editor. just replace the lyrics in the lyricss section and try to generate replacements. Trying to change 1 word can be iffy. If it's real bad maybe just try to cover the song with corrected lyrics and audio influence set high in the advanced options.
It completely changes the style of singing unfortunately when I try to replace the vocals in the studio 🙁
Studio could be different then the editor. Different ways to edit. It is a troublesome issue. When it is just a small like 1-5 second glitch, or mispronounce, can be harder to fix than a whole track.
Good afternoon everyone I have a question I hope someone can help me with. If I wanted to write a song and wanted to to have a minute Instrumental before the brige started back up how can I make that happen without it saying anything
Style boxes match. Adjusting sliders as needed. Trying to make selection that doesn't begin or end within lyrics. Start/Stop selection points should be inside of music when possible
thanks for the tips
let me try
If it's at the start of the song do an instrumetnal fist then extend. if want instrument later in the song add tags in the lyrics prompts like [Instrumental Breakdown], extend/shorten it as needed. Or create a new section inside an editor and then create instrumental in that. It should use the music before and after it to create a fill.
@frozen dagger thank you so much I will try that I appreciate your advice
i hope it helps. good luck
it's usually good practice to extend a track that isn't within lyrics, as I was saying. So, if you have your'e lyics prompt box end with [Instrumental], this is a great place for you to extend from. Start your extension prompt with the same tag [Instrumental] So it ends and begins with making music. Making it easier to lengthen, or shorten, that part as well.
was the multi-track editor a test feature no longer available? Editor now only shows one track with the full song rather than the stems
multi-track sounds like suno studio 🙂
Since the majority here are pro users, was wondering if somebody was able to create with versions after 4 a cover which sounds more or les the same as the original? With 3.5 was also able to create something like this as a free user, but with 4.5 it doesn't work. It could be that nothing useful can be generated without access to the advanced options. The audio influence value is set to 25% by default, and that's probably the maximum I can achieve with the free version, but that could just be a coincidence.
I've found it thank you - I believe the layout may have altered slightly. I had that section 'collapsed'
If covering using 4.0 (or 4.5) and sliders left at default, in what ways does the output differ than compare to doing the same with 3.5 on the same song? Can you list one or more points?
For istance, if vocals don't sound as similar, yea, that's the models defaults are different.
Something else? Tempo changes? Not replacing lyrics as well? More gibberish? Outputs tend to be shorter in comparison? Instruments change? Genre not as close to original (this one is true on 5.0) ?
I am free user so have no aces to that sliders.
Oh. Thought you were asking for advice that you've now upgraded. Ok. Nevermind.
With all covers the intent is that they cover quite closely.
As I have not logged into my Basic to check, I had just assumed sliders were made available for 4.5all only for all tiers, but I guess not?
If covering using 4.0 (or 4.5) 4.5-all and sliders left at default, in what ways does the output differ than compare to doing the same with 3.5 on the same song? Can you list one or more points?
For istance, if vocals don't sound as similar, yea, that's the models defaults are different.
Something else? Tempo changes? Not replacing lyrics as well? More gibberish? Outputs tend to be shorter in comparison? Instruments change? Genre not as close to original (this one is true on 5.0) ?
i see sliders on my basic
Have not upgraded, but hypotheticaly if I would was wondering if get beter results?
Well. What results are you getting with 4.5-all? That will be close to the answer. 4.5-all for Basic is the same for Pro and Premium. @frozen dagger notes that the sliders are available for Basic with 4.5-all, so it may work better that defaults to adjust the audio slider towards 75-98%. Others have gotten nice result, they say, but just leaving them at default.
My guess is that genres can play a part on how well it covers as well.
The results sound very diferent to the original.
I was starting it 30 seconds back it was just being a pain in the booty last night.
You my start with experimenting with the sliders to help match both song and genre for you. Otherwise, there's no other trick I know through starting with "Cover" that will consistenty produce a good copy across all songs, vocal characteristics and genres.
I can't use the sliders, since I am a free user.
For instance, only changing my style slider would change just the vocal timbre and leave the rest, basically, in the folk genre. Some things work that way.
Ok. Hold on. Ill try logging into Basic.
There are sliders under Advanced in Basic, for Weirdness and Styles. Try starting with no styles in the styles prompt box and adjust the style slider away from 50%, right or left.
Just in case you have styles posted, try the styles box with nothing, empty, when doing Covers.
I see them to, but they dont work for free users.
Haha... You know.... You're right! The cheeky little bastillians... 🤣 My bad for assuming again.
~~OK, third time then, to ask.
Describe a few specific change from 3.5 to 4.5-all you get with Covers, leaving the style box blank and having the exactly lyrics and arrangement in the lyric box.
Then maybe someone can pick one and suggest some sort of alternate trick.~~
Oh sheesh. I've been pretty slack these days. That's what working 15 hours a day get me.
I thought they only got rid of 2. 0 and 3.0, and kept 3.5 and added 4.5all. So, no 3.5 then either eh, for Basic tier?
It's in the announcement, I just didn't put two and two together.
Yea. So, you just have 4.5all and no sliders. The only suggestions is to get clever with style prompts and excludes that might steer the output back towards center to sound closer to the original.
I'm trying to create a song around my recorded vocals but they keep getting replaced by AI voice. Any ideas? I'm on V5 with influence set to 100%.
If you're trying to cover a song originally created in Suno, you might get better mileage from Remaster rather than Cover, as Cover typically tries to make variations on the original (hence the default audio influence being at 25%).
If you're not able to whack up the audio influence slider, I think you're going to struggle to get a replica in the new model, as you'll have to do it all through the style prompting.
try changing to "Add Instrumental" from the drop dowwn where it says cover in remix area.
Suno will typically replace any vocals you upload with an AI singer performing them.
If you want your voice to be used unchanged in the final version, you'd be better off creating a version you're happy with, exporting the instrument only stems, then adding your singing back over it via an external tool like Audacity.
Thanks for this, it's looking like this is what I'll do.
Yeah I've tried this about 4 times now and it always alters the vocals significantly.
if you have the option could try doing this in studio. add new track and then it will leave your vocal track unaltered, but create music on a new track. (don't write lyrics) (if you have vocal audio, dont put anything in lyrics almost no matter what anyway.)
in my small experience with the studio route it prolly going to create something that goes with the melody of the vocal uploaded, using the same timings etc, but if it creates a lackluster track. thats what i would take back into suno and cover.
I'll give this a go too. I like how the instrumentals match melody/timing just that they keep adding/replacing with AI singing.
Does anyone knows if the [Reprise]-Tag as section exist?
it probably wouldnt hurt to turn your vocal upload into a Persona and see what that does. but that is somethiing that is going to generate ai vocal to replace your upload everytime. not overlaying your vocal on music