📚┃suno-school

[finite folio]

I'm not the only zero here 😮

[mint granite]

I know the feeling, I'm a metalhead who loves composing soul/funk.

Here's my latest prompt that seems to work well:
Lyrics

[Tempo: 96bpm]
[Rhythm: uneven 13/8 (3 + 2 + 2 + 3 + 1 + 2), long-medium-medium-long-short-medium grouping]
[Vocals: Tenor female voice 44-year-old gritty, smoky]
[Slapped bass: Strandberg Boden Bass Standard 5]
[Guitar: Strandberg Boden Essential 6]
[Drums: standard pop]
[Mood: upbeat extatic]```

Style
```1990s slap groove funk, 96bpm, radio tube amp, alto female voice 44 years old gritty, no male voice

[finite folio]

I've tried adding verses to the song by doing a cover, sometimes it's successful, but not always.

[quick sonnet]

What you mean by successful? You just need to select the time, from where your track should be extended. You'll not always get what you like, but that's a diferent problem.

[hard beacon]

out of credits... now i have to wait 4 days for my 10.000 🙂

[finite folio]

I wanted to answer you, but I can't because I don't know how to write this answer in English and Google Translate is changing my original text into something else. Sorry. Sometimes AI does more harm than good, Google Translate is a great example of this. 🙁

[hard beacon]

use deepl.com instead of google translate 🙂

[quick sonnet]

No daily credits for Pro users?

[quick sonnet]

What's your language?

[hard beacon]

his language is "eesti" and baltic german

🙂

[quick sonnet]

As long it's not like German from Swizerland, should be posible to understand. 🙂

[hard beacon]

hahahaha

i think baltic german is much more complicated then swiss german

🙂

[finite folio]

This is even worse 🤣

[quick sonnet]

In Netherlands they have also a strange German. 🙂

[molten tulip]

Hi There, any one of you can help me please? I have genereated short radio jingle in Suno, I like it and want keep the whole song as it is- music, melody, lyrics, singign stlye except the language pronunciation what I want to fix. HOW HOW? 😄

[hard beacon]

remaster it 😉

[hard beacon]

coz it's not german ???? 😉

[finite folio]

Both of my great-grandparents on my father's side came here with the landlords during the time of the landlords, and one had the name I think ( Kießling ( in Estonia Kesling), of Bavarian origin, and the other had the name Ithink, of Saxon origin (Werst).

[quick sonnet]

If you are a Pro user, then do what Robin said.

[hard beacon]

😉

[molten tulip]

unfortunately that doesn't work well, Suno recreate the whole song with another melody, stly, the result would be just a new song, not same as original.

[quick sonnet]

Not tallking about Dutch, but German with that Netherlands acent.

[molten tulip]

yes I have subscription…I'm Pro, but SUNO not 😄

[hard beacon]

what is your problem??? you have created a "jingle" and you don't like the vocal-language-style???

sorry, but i don't get it 🙂

[finite folio]

Dutch is more similar to German than we think, it is written the way Germans usually pronounce words and there are also various words that are not in German. I think Dutch is a language that belonged to the Germanic family or was related to it.

[hard beacon]

dutch is a mixture out of english and "platt-deutsch" 🙂

[finite folio]

Very possible and sounds logical.

[quick sonnet]

Yes sounds like this, and many people there understand German, but understanding them, when they speak German is sometimes problematic.

[hard beacon]

ohhhh ya...

[molten tulip]

only the pronunciation of the language is wnet wrong, the style, melody everything in 8 sec is perfect, just pronunciation is trash.

[finite folio]

Mulle meeldib ajsu katsetada ja vahel see õnnestub kogemata. Vahel mul ikka veab ka, mitte ei lähe kõik alati valesti 🙂 ( This is in Estonian answer to you)

Looks like similar finnish?

😄

[hard beacon]

estonian is sooooo weird to read 🙂

[chilly wyvern]

Onhan se sukulaiskieli mutta en ymmärrä kaikkea.

[quick sonnet]

Don't speak Suomy. 🙂

[hard beacon]

am Ende des Tages ist deutsche Sprache doch einfach viel schöner 😉

😄

[finite folio]

its actually very simple to read, long vocals you read like in German ah , eh, oh, ih, uh, äh,öh 🙂

[hard beacon]

😄

اللغة متعددة الجوانب

;-D

[quick sonnet]

But people still say: "Deutsche Sprache, schwere Sprache". 🙂

[hard beacon]

absolute

[finite folio]

"Night" in Estonian have only double letters "öö" you can read this like "öh" 😄

[hard beacon]

german: der, die, das.. english: the

😄

[quick sonnet]

Gues you had no French? 🙂

[hard beacon]

I had it at school in 1997... but only for a year... I failed completely.

[finite folio]

in Estonian does not exist like 2 diferences genders like in Greman or in English have german er,sie and English Hi , she estonians say that just one word Tema or ta ( ta is short vorm of tema)

[hard beacon]

i just know: va te faire foutre 😄

[quick sonnet]

Had just "Grundkurs", but helped me at University.

[hard beacon]

i had nederlands on the univeristy for a year.. but my lnowlege is gone...

*knowledge

[quick sonnet]

Do you mean Dutch?

[hard beacon]

i am to old to learn new languages 😉

ya.. dutch

Ik ben te oud om nieuwe talen te leren 🙂

en grammatica is slecht

dutch: slecht / german: schlecht

🙂

[quick sonnet]

If I remember right, you just about my age, so not really old.

[hard beacon]

let's test it.. what's your age ??

[hard beacon]

ok.. 43

🙂

your brain is better then mine 😉

[finite folio]

And Estonian does not have articles like in German or English "der, die, das or The". Estonian has prepositions, adverbs and conjunctions and in addition Estonian has 14 cases for one noun, including the noun in total 14. for example

[quick sonnet]

And got answers to all questions. 🙂 Gues you know where that quote comes from.

[hard beacon]

sorry, but i am lost at this point

😄

[finite folio]

What this test give ?

[quick sonnet]

https://simple.wikipedia.org/wiki/42_(answer)

[hard beacon]

the test gives, that zero and i have chatted before... he can remember but i can't 😉

[finite folio]

I know another quote. Dutch is a false German language.

[quick sonnet]

Don't need to remember, can see in DMs.

[hard beacon]

what the...

oh...

😄 hahahahaha

[finite folio]

I'm 50, what does that mean, according to this book :D, that I shouldn't exist anymore? 🤣

[hard beacon]

50 is the new 20 🙂

[quick sonnet]

You have not watched that movy, right? 🙂

[finite folio]

But let's assume that 0 is not a number 😄

[quick sonnet]

Just not a natural one. 🙂

[finite folio]

Actually, I'm a big sci- fi fan, but I've not read this book and I don't think I've ever seen the movie.

[finite folio]

Just not natural infinity? 😁

[quick sonnet]

Book? Wasn't the movy based on a game?

[quick sonnet]

But without 0 modern AIs wouldn't exist, the same for 1. 🙂

[finite folio]

There are two infinities in numbers, 0 and 8. 🙂

[quick sonnet]

Binary system knows no infinity. 🙂

[finite folio]

But a long time ago I read some strange article that claimed that the original number sequence ended with the numbers 0 to 8 and that when people reached an This 9 was a fairly new number and related to the emergence of mathematics. I don't believe it and I can't argue with it.

[quick sonnet]

Are you tallking about octal system, where 0=8?

[finite folio]

maybe, I've never excelled in math 😄

[finite folio]

But I don't use a lot of math, I get the same answers purely as a result of logic, without doing formulas or calculations 😄

[dusty panther]

came here to maybe learn a thing or two about prompts, learned facts about european languages instead

[torn spoke]

the pain

[lyric kestrel]

anyone have an idea for a prompt? https://www.youtube.com/watch?v=pcdy_KCKZ8U

[quick sonnet]

A prompt for exactly this song?

[finite folio]

Good night 🙂

[lyric kestrel]

yeah

[quick sonnet]

The simplest way is to upload the song and let suno generate the corresponding prompt.

[lyric kestrel]

I have but it just makes a bad prompt which sounds not anything like it (obviously not the same but yknow)

[sour laurel]

i extracted the stems and they're visible in the studio as well as in the song list, but i when i click "extract stems", there's no see extracted stems buttom, therefore i cannot download the stems in MIDI or WAV bc when downloading directly from the studio u can only download mp3, can anyone help?

[frozen dagger]

checked your "My Workspace" Folder?

[sour laurel]

the stems are there. problem is, i cannot download the midi folder, which afaik is only possible through that button "see extracted stems"

[arctic elk]

No problem, it's all good...

[sour laurel]

and if not, it would be awesome if anyone knew the way, since i spent 850 credits extracting stems from various songs, specifically to download the midi files

[frozen dagger]

afaict... there is no midi function directly? dunno homie. i've heard people converting stems to midi, but i dont knwo that they used Suno to convert. Im asking gpt to try and help ya out. its just as confused on the Midi part. closest thing it coudl find was The knowledge-base article for Suno clearly states: “Stem Extraction … you can Download any individual stem, Download All, Regenerate, or listen …” with no mention of MIDI export.

[sour laurel]

yeah u can choose the format in which you want to download the stems and it turns it into a zip file

[arctic elk]

This worked for me. Hope it helps: Prompt:

Emotional male + female duet — cinematic yadda yadd yadda your prompt style....
Production Notes for Suno

Mood tags: duet, cinematic, emotional, male + female vocals, orchestral pop ballad, passionate, heartfelt.

Instrumentation: piano, strings, soft percussion build, acoustic guitar layer near chorus, reverb-heavy ambience.

Dynamics:

Verse 1: solo male, quiet.

Verse 2: solo female, echoing tone.

Bridge: call & response buildup.

Final chorus: full orchestral crescendo.

Outro: soft fade, reverb tail on vocals.

Lyrics (Duet Script):

[Verse 1 – Male]
Your lyrics...

Your lyrics...

[Verse 2 – Female]
Your lyrics...

[Pre-Chorus – Male → Female Echo]
Male: Your lyrics...
Female: Your lyrics...
Male: Your lyrics...
Both: Your lyrics...

[Chorus – Together / Harmony]
Your lyrics...

[Bridge – Call & Response]
Male: Your lyrics...
Female: Your lyrics...
Male: Your lyrics...
Female: Your lyrics...

[Chorus – Together, Crescendo]
Your lyrics...

[Outro – Soft, Unison]
Your lyrics...

[sour laurel]

the problem is, some songs which i extracted the stems from do not have the button to see the extracted stems

[frozen dagger]

i see it, trying it out myself see if can still assist. blind leading the blind. woo

[half dragon]

Yeah, I noticed that too - that if you extracted Stems in Studio, only way to see them Stems are to open it again in Studio; no button.

[frozen dagger]

ya, i was in the create tab, got all stems downloaded both version's midi.zip. the stems only appear when the song track is selected in studio. usually id see those stems in my workspace, but i do not, only inside Studio can i see the seperated stems from the track.

[frozen dagger]

mny bad they are there

[frozen dagger]

i had to uncheck hid stems... duh

[half dragon]

oh yeah in Libraries you always do see those stems clips but kinda hard to tell which they're from.

But I remember if you extracted outside of Studio, there'll be a link to get to the Extracted Stems and they also show up in Studio.

[frozen dagger]

all i saw was the button at the bottom says edit in Studio and it just placed the stems on the timeline in Studio. Not sure what button to show the stems ya'll were refering too. sounds like some feature was removed? they have been saying V5 is base for Studio "and all to come" so maybe part of transistion to Studio then.

[half dragon]

hold on ... let me check

[frozen dagger]

if u want the stems, uncheck the "hide stems" box in filters, and you can still download the stems as mp3 or wav. midi zip only from that 1 "extract stems" screen, looks like after that is closed have to sspeend credits to get midi again. sorry, not more help. hope u get it figured out.

[half dragon]

This is what I meant. You can get back to that screen to see them Stems if you extracted outside of Studio.

[frozen dagger]

i see, said the blind man

[azure crane]

Anyone know how to get Suno to not change tempo in the middle of a song and just get faster and faster... This was a problem before but now it's back... So frustrating.

[frozen dagger]

listing any genres of music will have it's own affects. try to guide it with things like [Consistent xxBPM throughout] [Metronomic] [Steady Tempo] [Sempre tempo giusto] [Senza rubato] -“without flexibility,” literally “without stealing time.”

[Non-rubato]

[azure crane]

Ok Will try, thanks for the tip 🙏

[frozen dagger]

gl. style genres have their own associations if you having problems try altering them, if u have, instrumental info. Just saying, if you want a steady tempo and you are working with something with like Jazz Ballad or an Opera, something naturally rubato its could choose to pull from that instead.

[frozen dagger]

i lsited them like tags but you may want to put that stuff in your style prompt, or both

[azure crane]

Well, it was just one track today that hade that tempo change. Tried so hard with covers to make it in one tempo during the whole track but it failed every time.

[frozen dagger]

adjusted your sliders? (what a like Suno-dad thing to say)

[azure crane]

Tried everything… gonna try your tips now…

[frozen dagger]

https://tenor.com/view/dr-seuss-grinch-gif-20631463

[polar spade]

Oops. What'd'I do

[paper dawn]

Anyone using studio to create there songs track by track and having issues just remember to set your each track to original tempo instead of being on beat and it’ll come out nice, each track is for individual sounds like bass, synth, drums etc

[vale isle]

problem: using studio to add vocal track, and it keeps adding background instruments.

[paper dawn]

Do you have to tell it a vocal track only no instruments no drums no percussions etc. I usually go to a different Suno tab in my browser and I’ll type up what I wanted to do and then hit that little button that changes it to the right way of typing it

Yeah if you and then you have to make sure it’s on vocals

If you don’t do all three of those you’re most likely gonna get some sort of instrumentation

[frozen dagger]

didnt see this before. glad you shared. thanks.

[paper dawn]

No problem brother I’m glad it helped

[azure crane]

Nothing helps... Every Cover I do now which started today changes the tempo and getting faster and faster...

[frozen dagger]

using v5? try other models?

[azure crane]

Nope, cause v5 is totally superior for my musc.

Tried 4.5+ it the tempo was steady as a rock... Something is wrong with V5...

[frozen dagger]

v5 is more reliable for sound quality, but it gets that reliability by not being as creative as earlier models. v5 songs feel more like templates. imo.

[paper dawn]

Maybe there’s something in your prompt do you have any thing extra in [verse] like [verse - building layers] I found most of the time those things never work I do them all separately but just asking cause that’s when I get that experience is usually it’s in my tags

[azure crane]

Well this is my last day of the subscription period and this is the only day this has happened so something is not right with the V5 today….

NoNo, it has nothing to do with my Lytic section cause Iäm creating only Instrumental music...

[paper dawn]

Hmmm what’s your prompt say?

[bronze harbor]

Just spent 30 credits and got back nothing but silence. This isn't the first time. This happen to anyone else?

[paper dawn]

Yes they usually refund those credits you spend on those

[bronze harbor]

oh ok so I just write to support about it?

[paper dawn]

Its an extend ?

[bronze harbor]

me? no it's in "replace/recreate" in Studio

[paper dawn]

Ohh and you’re under the take lane part where it shows the new music that?

[bronze harbor]

yeah, all 6 takes are just silence.

[paper dawn]

Ohh yeah I never experienced that but actually right now I’ve noticed like two or three different people are saying something’s going on with Suno so I say is a maintenance but yeah you could definitely send a message to support ask him what’s up with that

[wise kindle]

what's the best way to minimize the intro time? I feel like 60 seconds of Oohs, Aahs and guitar solos is just way too much. obviously it's probably best to avoid terms like "short intro" because "short" to AI might mean 45 seconds which is still too long.

[frozen dagger]

plus it will see "intro" and that tag itself can create what you are trying to avoid. try avoiding the [Intro] tag completely if want shorter intros. that would be where id start. dunno what you got so far, but that may be part of the issue if you have that.

[wise kindle]

Nah I know AI can get confused with negative words you try to use in prompts, like "short <something>" or "no <something>", so I haven't supplied anything for the intro in there at all currently

[slow crag]

has the website bugged out cant create anything?

[wise kindle]

just created something a minute ago

[slow crag]

not working for me for past hour

[wise kindle]

sounds like a you issue, sorry

[hollow ridge]

Cant too

[daring dune]

How would you go about not including an instrument. I want to generate a solo singer with a solo instrument and it still adds other instruments no matter how I try to formulate myself.

[raw oasis]

key word acapella?

[daring dune]

I would think that is would be sans the instruments but I will give it a try. thanks

[raw oasis]

or if you mean "only a vocalist and a violin, no other instruments" for example could try that.

[daring dune]

Yes, I want a vocalist and a harp

No other instruments seem to work 1 out of 10

[raw oasis]

https://suno.com/s/FjHOaI1yGHMiq17I

"solo vocalist with harp only. sparse and quiet single instrument."

[daring dune]

Cheers.

https://suno.com/persona/eff132ac-c496-4162-8892-94cc5533dbe9

This is what I have managed now with a few too many generations

[dusty panther]

anybody figured out how to disable the "ooOoo" intro that suno put everysong

[raw oasis]

actually that will be perfect for my short film I am starting. my solo guitar player can do the song this way... later it will be full production.

[daring dune]

Harp only didn't work.

[raw oasis]

do you mean when you don't have a singer?
Otherwise 2 out of 2 for me got single instrument and singer w/ "solo vocalist with harp only. sparse and quiet single instrument."
https://suno.com/s/IWtHa6Y5EuwOfVRY

Will try acousitc guitar now.

[daring dune]

No, 'harp only' didnt make SUNO think that there was one instrument only in the song.

[raw oasis]

I just love Suno. it is brilliant. never thought I would see anything like it in my life.
https://suno.com/s/oshednVb4IQdbSBH

[daring dune]

My guess is that the style of song it wants to emulate has its own set of sounds that it needs, or at least feels compelled, to use.

I agree. I have made some hautingly beautiful songs. Would love if it didn't take 20 itterations of each to find one that it nails 🙂

[raw oasis]

Oh but thank goodness it does or we would have 20 million hit songs out there.

I fear it will get so good.. music will become blah.

[daring dune]

No, hit songs it nails every time. its the songs that breaks the mold that is struggles with

[raw oasis]

oh yes, like my Broadway/Prairie stuff.. yeah that gets complex. Are you using Editor/Personas already for that kind of stuff?

[daring dune]

Im working on a persona yes. But it is one instrumentalist that sings and SUNO loves to add artists to the solo act 🙂

[raw oasis]

not experiencing that. I must be confused. Got 4 out of 4 with a singer and 1 instrument.

[daring dune]

Its very genre dependent

Im doing celic/welsh folk songs on harp

[raw oasis]

but I did not put any genre prompts in my song.

okay so I will put welsh folk song in mine and see. My ancestors were from Wales 😄 like 1500..

ugh.. added tiny bit of Piano that time.

[daring dune]

Yes, and it loves to put some flute into the songs as well 🙂

Sounds great but I really wanted the solo player experience 🙂

[raw oasis]

she got a little bit of a flare of accent ya know..;) https://suno.com/s/qGgI1NuXrEczxSa1
"solo vocalist with acoustic guitar only. sparse and quiet single instrument. Welsh folk."

[daring dune]

You linked that before. She probably did it my way and just itterated it until she got one that was solo.

[raw oasis]

No. that is a brand new generation.

anyway.. were starting to repeat. Suno is brilliant. Keep at it! ❤️

[daring dune]

Thanks for the engagement friend. Ill keep working on the wordings.

[raw oasis]

YW - oh also. try some celtic scales possibly. Suno will respond to chords, keys, modes sometimes.

[daring dune]

I havent even started there yet. thanks for the suggestion.

[raw oasis]

Modes for Celtic music
Folk music from the Celtic tradition generally only uses four of them (apart from some modern tunes) so it is on these four that I will focus in this blog, but as I’ve said it will be VERY useful to you to remember the names and orders of all of them.

The four Celtic modes are:

Ionian – standard major scale
Dorian – “optimistic” minor scale
Mixolydian – major scale with a flatted 7th note (think blues)
Aeolian – “pessimistic” minor scale

[restive pebble]

ahh the diatonc modes, very useful in vocals

[lament fern]

Hi, how to say that I want each text line for 2 bars, even the text line is shorter than 2 bars. Actually, suno use 1 text line by bar, it's too quick for the lyrics. Thanks

[restive pebble]

I do not believe you can xplicitly prompt suno into following bars. Try using [break] or putting in an [interlude] between

[quiet solar]

Not experimented with this personally, but you could try [1 bar interlude] breaks. Artificially splitting the lyrics into two lines should help, but you might need a few tries to get the cadence right.

[restive pebble]

Sounds like it'll work. I will try too

[near flume]

Suno doesn't understand time or bars... yet.

[lament fern]

So, How can I ask a break before the next line ?

[restive pebble]

[Break] [Pause] but break is more effective.
You can alsotry using the elipse ... as it sometimes works

[formal robin]

Hey, quick question — I’ve been trying to get Suno to start a track with a clean kick and snare breakbeat intro, like a DJ-style lead-in before the main elements drop. No matter how I prompt it, it usually starts with vocals or random sounds instead of an actual beat. I’ve tried different wording like ‘start with drums only’ or ‘DJ intro with just kick and snare,’ but it never seems to stick. Is there a reliable way to prompt or structure the intro so it begins with a drum pattern for easier DJ mixing?

[hard beacon]

V5 doesn't like extended mix Intros. Try it with v4.5+ and use remaster

[wanton cradle]

do yall make your song thumbnails in 3:4 or 1:1?

[warm jolt]

What do I need to put in the lyrics if I want to extend an instrumental 30s intro to a full instrumental song? Suno only extends +19s or so over and over again. Then the song is less than a minute.

[quiet solar]

Try making the intro a persona, then create a new instrumental song using that persona with Lyrics removed and the style prompt saying that it's instrumental, and describing what you want.

[young fox]

How do you make the song or voice sounds like hatsune miku

[warm fractal]

Does anyone know a bracketed instruction to make the vocals of a section sound very emotional, or voice breaking? Can't seem to find the right command to do it. (If it's even possible.)

[warm jolt]

Okay I'll try this. Seems like a workaround though 😅

[quiet solar]

@warm jolt Lots of "I want to do (xyz...) in Suno reliably..." require what some people might call workarounds.

[quiet solar]

To the point of a voice breaking down, no.

You could try "passionate" as an adjective. Alternatively, look up some synonyms for passionate or emotional and try those.

[restive pebble]

Haven't done many things without lyrics as I am a writer but when I want a long intro I always use some descriptive text in the styles section referencing the intro as an instrumental, then I us the style tag in the lyrics field above my lyrics. [Intro: Instrumental] Not always but occasionally it will give me up to a minute of music and adlibs before the lyrics begin.

you can add more to the intro tag for specifics if needed. Drums Synth etc

[steep haven]

I created a persona for my virtual band, based on a pre-existing track, of course. This is very useful to ensure that the voice and overall style are always the same, like in a real band. The problem: Suno ALMOST ALWAYS copies the intro and/or riffs from the original track in the new tracks I try to create. I want them to sound like the same band, but not always like the same song. Does anyone know how to fix this? Thank you.

[last bay]

depends on audio influence but you can try making persona from voice stem and try style more without persona

can also experiment how it sounds when using the song from persona as inspo instead and then cover

[frail marlin]

Anyone knows how to solve latency in studio when I'm recording voice channel ? And how to improve the sound ?

[thick gust]

Anyone has idea of a style that could help me create an egyptian like vibe? Because i tried Shaabi or traditional egyptian and it keeps doing some arabian stuff or drops heavy guitars coimpletely without wanted vibe...

im out of ideas at this point

[frozen dagger]

create a new track without the persona first, then extend from with persona?
"workarounds" >.>

[slow crag]

I did a trance song with just using country style it sounded middle eastern

[frozen dagger]

any examples of known artist to compare what you r looking for?

[thick gust]

https://www.youtube.com/watch?v=vslsS-Uu5x4&list=RDvslsS-Uu5x4&start_radio=1 maybe something like this could help

[slow crag]

@thick gust Genre: Epic Cinematic Hip-Hop with Middle Eastern instrumentation

Mood: Majestic, commanding, mysterious
Instruments: Oud, darbuka, deep bass, orchestral strings, ancient flutes, trap beats
Vocal Style: Deep male vocals with a commanding tone, chant-like delivery, layered with echo effects
Lyrics Theme: A powerful pharaoh rising from the sands, commanding the Nile, building pyramids, and ruling with divine authority. References to Ra, Anubis, and eternal legacy.

This might help

[thick gust]

thanks, i'll try

[slow crag]

Here's an example

[outer spade]

I'm working on a project and trying to modify a short Gospel choir section. I'm using the tools within Suno Studio to enhance it, but every time I try to 'improve' or convert the track to a different Gospel style, the generated output completely changes the original melody.

The issue is that Suno is applying great improvements, but with an unwanted melodic variable. I need the enhancements without altering the core melodic line. Has anyone else encountered this problem when modifying vocals/choirs within a Suno project? Any tips on how to force melodic consistency?"

[slow crag]

https://suno.com/s/3aYO4QfkdoinWJ4h

Might just take out hiphop make it more authentic

[frozen dagger]

when i saw ancient egyptian instrumental created bland generic middle east music, talked to gpt and i came up with this. just mentioning Ra and Egyptian things is going to have the ai start looking there likely. but this was my attempt:
Ancient ceremonial music evoking lost desert temples and golden tomb halls. Slow hypnotic pulse around 85 BPM in a pentatonic minor mode. Deep clay hand drums and frame percussion form a trance-like rhythm, joined by wooden clappers and metallic sistrum rattles shimmering in echo. Layered reed flutes and double pipes drift in microtonal harmony, answered by plucked harp and lyre drones resonating through stone chambers. Ethereal female voices chant wordless invocations with airy vibrato and melismatic phrasing, rising and falling like desert wind. The mix feels spacious, mystical, and reverent — distant echoes, limestone reverb, and dust-filled air. Cinematic tone, sacred, ancient, transcendent — a ritual soundscape of divine balance and eternal dawn.

https://suno.com/s/nySzP2qLNYN4kTCZ

[thick gust]

time to dig into stuff like this one

[slow crag]

Did it work for you?

[thick gust]

suno is cracking up atm

generating me 7:59 songs without any music

:v

i'll try to refresh and try again later, idk whats happening

[thick gust]

Yh, it makes me some trap/hip hop stuff even tho i removed it

[slow crag]

That's a bummer

[thick gust]

i repoted them all but idk if it will give me back points so

i'll try to mix Shaabi, egyptian, traditional egyptian together or something

[slow crag]

@thick gust Genre: Ancient Egyptian ceremonial music

Mood: Sacred, regal, mystical, timeless
Instruments: Ney flute, sistrum, lyre, harp, arghul, frame drums (light, slow rhythms), ambient desert wind textures
Vocal Style: Chanting male and female voices in ancient Egyptian or stylized vocalizations (no modern language), echoing temple acoustics
Structure: Slow, meditative progression with long pauses and droning tones. No modern beat or rhythm — purely atmospheric and ritualistic.
Just take the vocals out might work

[frail marlin]

How do u work on the sound of a voice track that I recorded on Suno studio ?

[thick gust]

By the way do you know a some tips or tricks how to write if i want to make some verse to each line be sang by different voice

[Male Vocal]
text
[Female vocal]
text

It keeps screwing it

[slow crag]

dont you put vocal style; and put the style your after in it

[thick gust]

[slow crag]

what kind of style are you after and i can try and look it up

[thick gust]

I mean its not style, i just need a tip if there is any to make the bot sing female and male vocals in a verse and each line have a different one like

im trying to make the bot sing line by line each on different gender vocal

[slow crag]

aww like making a duet?

[thick gust]

yeah

and each line of text different

mixing the verse

[slow crag]

one sec i can find out for you

[slow crag]

🎤 Step-by-Step Guide to Suno Duet Prompts

1. Use Character Tags in Lyrics

• Begin each vocal part with a name in brackets to cue Suno to switch voices.
• Example:
  [Alex] I saw the sunrise over the bay
  [Jamie] And I felt the warmth wash fears away
  [Alex & Jamie] Together we’ll chase the light today
• This method works across genres—Jazz, R&B, Rock, Neo Soul, etc.

2. Keep the Format Consistent

• Always use the same bracket style (e.g., [Name]) and place it before each line or section.
• If you want harmonies or overlapping vocals, use “[Name & Name]” to indicate a shared part.

3. Describe the Duet in the Prompt

• In your Suno prompt (the description field), mention that it’s a duet.
• Example: “A heartfelt acoustic duet between two lovers, Alex and Jamie, with alternating verses and a shared chorus.”

4. Choose Complementary Genres

• Duets work well in styles that emphasize vocals and emotion: ballads, acoustic pop, soul, folk, etc.
• You can experiment with contrasting genres too—like a rap verse paired with a sung chorus.

5. Add Emotional or Narrative Cues

• Suno responds well to storytelling. Describe the relationship or situation between the two voices.

so [female name] [male name] on each line you want them to sing

[thick gust]

does it work better than putting female or male vocal?

[slow crag]

i would have both so it works fluid unless you want to use a single person to sing it

well bot to sing it 🙂

[thick gust]

ill try names

[slow crag]

😄 happy to help

[thick gust]

thanks

[frozen dagger]

getting male and female vocals when and where you want them is on top of list of everyone trying to get something that works consistently, with text only. If you wanting it to be specific then covering the song with the opposite gender and separating the stems and then splicing together is the only really guaranteed way at this time.
[Call and Response] and puting your tag can help but it is never 100%

[vague sleet]

I am doing crops for preview music videos. Now i am facing a problem from suno where i Crop a song, play it back and i find that the vocals are low quality like the sound that made it pop has disapeared. I know this is problem that is only on suno, because i downloaded the crop and it was the original sound i need. but listening to the crop through suno it has become ruined. I don't know what to do. Please fix this issue.

[stable oasis]

Is there a way to specify an instrument or sound effect in a specific place in a song, such as for an instrumental bridge?

[mild egret]

[Interlude: sax solo, 12 bars]

[mild egret]

Sound effects a bit hit and miss. Some suggestions on my link (click my face for the link). More than half-way down, "13. Sound Effects"

[stable oasis]

Thanks!

[graceful condor]

Man they really need to work on the extends especially with complex rock music. They are terrible like all over the place. Drums totally absurd, structures without sense and logic regardless how you tweak and turn the sliders etc. It is really frustrating.

[crude zinc]

What a solid way to nail duets? I tried Male voice 1 and Female Voice 2 tags/prompts

[quiet solar]

There is no solid way to nail duets, but this topic comes up about 5 times a day, so if you do a search for duets you'll find the best advice available pretty easily

[crude zinc]

Lol thank you

[quiet solar]

@crude zinc And to save you a bit of time:

If you start from a clean generation, specify both vocalists separately in the style prompt and tag throughout the lyrics, then put up Style Influence to 80%+ and Weirdness down to low-ish levels (e.g. 30% or below), it should be more likely to work. Note: "more likely", not "always will".

The more times you switch between voices, the more likely it is there'll be a mis-switch. The tag [Call and Response] is helpful if the switch happens line by line, but incredibly unhelpful if it doesn't.

And if you're switching over mid-verse, you might want to think about how you actually lay out the verse in the lyrics to help Suno find and follow the pattern. In the example, you'll see there are deliberate line breaks between the male/female sections in the core of the verse, then an arbitrary [Pre-Chorus] heading specifically to separate out a section from the Verse where the original creator wanted to change the pattern.

It's very unusual for Suno to get an entire song right first time with vocal switching, but if you are determined and happy to burn credits there are a few ways it can be done within Suno, and if you're comfortable with using a DAW or external audio editor then there are a few more options (which I personally prefer).

[quiet solar]

I've not done it specifically with choir vocals, but depending on what your enhancements are, have you tried downloading the audio of the section you're trying to adapt, re-uploading as an Audio prompt and then remastering it?

[late mason]

Hey does someone have a good prompt to make songs like Juice WRLD ? Lucid Dreams ??

[steep haven]

I thought about that, but in that case the generated track would have a different voice than that of my "vocalist". 🤔

[frozen dagger]

not if you generate an instrumental. otherwise, yes, that is correct. you would have to add another work around to crop the intro and save it as a new song with no style (or the style you want for the song its to be used for), as a new file just to remove the vocals/data you didnt want.

[keen iris]

I have The Question: why v5 makes only 5min tracks?
Even when I trying to cover longer tracks (6-7 mins) it stops at 5:11 and it's not the end of the track so I need to initiate "Extend" feuture wich is very bad in v5 so I never get what I want with long tracks.
I spend more than 1500 credits (on one track) only for extends and covers and still not getting final results, I'm very upset with that.
v4 was making over 8min tracks well, why v5 can't?

[hollow hare]

https://discord.com/channels/1069381916492562582/1422252931943501935

[wise kindle]

if your song is over 4 min I probably ain't listening to it anyway

[keen iris]

It's ok, the more exp you get in music tree the longer and harder music you'll handle 😉

[slow crag]

What are good combinations for hybrid music

[wise kindle]

my favourite this week is electronic + hair metal

[slow crag]

I've been dabbling with Middle east and trance style techno getting bored now will try that

[wise kindle]

been converting some of my original chiptune tracks to electronic, hair metal and it's going really well. hope it works for you!

[frozen dagger]

try experimenting with weird/uncommon instruments too. fun stuff.

[wise kindle]

electronic, Scale: Freygish/Phrygian Dominant, melodic, rhythmic, punchy beat, hair metal, through-composed form, clean mix, clear production quality
is the full prompt I've been using. you can probably omit the mix stuff. I only added that to try and get a cleaner sound seems I was covering pure chiptune tracks with square waves and white noise drums :P

if you're sick of middle eastern stuff probably drop the scale too lol

phrygian dominant has that middle eastern sound to it

[spring veldt]

Is there a way to get the same vocals on a number of songs or do I just have to try my best to generate something similar with my prompts?

[slow crag]

I just made a song with electropunk and orchestra

https://suno.com/s/bk786hLcApi3ugje

[frozen dagger]

separate vocal stems, use them as personas

[spring veldt]

ty makes sense. I’ll give that a try. I did this using a recording of my own voice (which isn’t good! 😭) so I should’ve thought about doing that with a vocal stem

[slow crag]

I'm gonna start keeping persona's i forget to keep them

[frozen dagger]

Use Cover on your vocal. give your own sonic identity a little pizzaz -> then turn it into a persona

[crude zinc]

Almost nailed down the duet https://suno.com/s/OBlc2YO70lrz6Dxe

[frail marlin]

How can I edit the sound track I recorded on Suno studio ?

[half dragon]

What kind of edit?

[frail marlin]

Just sound edit

[half dragon]

I think you want to use Edit tool?
From Studio:

1. select the clip
2. 3 dot menu
3. Remix/Edit
4. Open in Editor

[frail marlin]

Ok thanks I'll try it

[half dragon]

...but you can do the same in Studio... select the part you want to fix... and add a new track, then regenerate

[frail marlin]

If I like the sound of the voice in the song that Suno made - can I install my voice recording in this track so it will get the same sound ?

The same edit I mean

I want my sound to have the same edit but to keep my sound. That it will sound me

[half dragon]

I don't think that's going to work in Studio
There are a few options but will require some extra work:

• Upload your voice first, Cover it to create a song, then create Persona
• Then with the existing song,that you want to edit, Cover it and apply that Persona; play with the slider settings so it would keep the melody and arrangement as close as possible
  Another way is to swap out the voice off line using 3rd party software or spit the stems, download the instrumental track then sing over it.

[frail marlin]

Ok, thanks. This is something Suno has to improve. Why all this hassle ? Also, I recorded my voice and there was latency that was very hard to fix. Took me an hour. Is there a way to fix this latency ?

[half dragon]

Also keep in mind, your uploaded voice used in the song would be as close as yours but would never be the same.
Suno won't clone the voice.

[half dragon]

What kind of latency? timing with the instrumental track?
record small section and then line it up with the beat.
Or download the instrumental track and record off line - then upload to Studio

[frail marlin]

Ok thanks 🙏

[raven turtle]

I’m finding success with this by just “reuse style” for another track. I’m getting the same vocalist on each one and have built up about two albums worth of good tracks this way that sound consistent

[regal hamlet]

Hello everyone, dont know if this is the right place to ask , but i had more than 9000 credits left now all of a sudden its all gone

anyone know how to solve this problem ?

[restive pebble]

Wow, sorry to hear that. I think this is the right place to ask for help from the suno staff.

https://discord.com/channels/1069381916492562582/1372994553233276970

[regal hamlet]

ah thank you very much ! 😄

[restive pebble]

for the same vocals, use Personas. They seem to provide the most consistent vocals for me.

[gusty pendant]

Hello, is there any good youtube videos showcasing working in suno studio? Would help a lot thanks.

[wet osprey]

Same thing happened to me.

[graceful condor]

Comparison of Progressive Metal or 'Djent' - compared to any other genre is enough to make me abandon prompting for it altogether. I tried to EQ it in DAW, then load the bounced track back into my DAW. Even with a Limiter and volume jacked up, it still sounds horrible compared to the Drum and Bass song that I haven't done anything to. Anything heavier than rock and roll turns into a band in a sewer pipe quality or it's either so harsh with low end or high-end frequencies that hurt my ears. Suno can't hang onto the Death Metal genre prompts very well. It's the only genre that I have problems with. Everything else turns out fine with minimal effort in positive and negative prompt. Brutal Death Metal sounds okay, Technical Death Metal-Melodic Metal sounds okay. There's too much going on in this genre for Suno to handle. Why?

Seeing the audio volume difference of these two tracks in decibels makes no sense that the drum n bass song is louder with more clarity and pleasant texture even at a lower decibel. I can raise the DNB song volume and still not have to use a limiter where the metal song is at max, if I raise the decibel any higher, I'll lose more frequency. I think it's time for me to hang it up with Suno metal.

[graceful condor]

Since Microsoft ended their support for windows 10 on October the 25th weird stuff has been happening even with my ISP. That was Saturday, right after Suno crashed, Amazon went down right before that. Since Friday I've been unable to access some of the websites that I normally used. My Suno account is bugged now. I blame Microsoft, hiring people that aren't qualified.

[finite folio]

Hello

I think i need alitlle bit help, yes i know again 😄

I want to add vocals to an instrumental song that I have previously prepared on my computer with Mixcraft 10 Pro Studio, so that a female alto voice sings a soft "oh-oh-o-oh" and immediately after that a male voice sings a raspy voice "och-och" or something like that. How should I mark this and where so that the AI ​​can understand it?

I've tried several different variations, added extra lyrics in a separate lyrics box also, and marked the style, but it doesn't work. AI doesn't do that. Sometimes AI adds a female voice, but not a male voice.

[polar spade]

@willow jasper asks: ".... Does anyone know if the Suno Pro commercial rights includes the ability to play the music in a public retail shop environment. . .?"

That's a bit open-ended. Please explain what you're trying to do. Play your own music in your own shop?

[half lintel]

How do I cancel the subscription

[finite folio]

Maybe some DJ who mixes music from well-known songs and plays them in clubs or at larger events can answer this? If there are any here of course, but maybe you'll be lucky 🙂

[polar spade]

Your subscription page has a Cancel Subscription button to downgrade your subscription. Not completely remove your account.

[finite folio]

But I still can't add an instrumental to the song, which I wanted. What am I doing wrong 🙁

AI don't understand what I want, I've tried different options several times, but to no avail.

[polar spade]

I need to know what details (scenario) their question is referring. In the USA there are copyrights laws regarding public play.

[willow jasper]

my bad, yes I intend to generate music using Suno Pro & play that music in the background in my own shop in the UK. I’ve looked through the guidelines and tos and can’t see anything that explicitly states that’s okay

[finite folio]

"What is commercial or non-commercial use?" Can this place in Suno's help provide an answer to this question? Couldn't post the link 🙁

[polar spade]

It's a bit tricky since UK law require shops to have a public license for ALL public music played. So even though you're not going to sue yourself for infringement and non-payment of royalties, the license itself is not contengent on whose songs you're playing.

Nope, I take that back... Unless you hold ALL rights. Then you're likely fine - - you should fit the exemption.

[finite folio]

I want to add vocals to an instrumental song that I have previously prepared on my computer with Mixcraft 10 Pro Studio, so that a female alto voice sings a soft "oh-oh-o-oh" and immediately after that a male voice sings a raspy voice "och-och" or something like that. How should I mark this and where so that the AI ​​can understand it?But I still can't add an instrumental to the song, which I wanted. What am I doing wrong.
AI don't understand what I want, I've tried different options several times, but to no avail

[polar spade]

@finite folio @willow jasper @Any public performance, where anyone else but staff can hear, including just playing radio broadcasts, typically is deemed to require a public license and admittance of royalties paid to the proper Performance Rights Organization or collection society.

The crux is keeping proper records and able to prove that the music you've been playing for 6 months does not require payment of any royalties (since they are allegedly were your songs) to the PRS or PPL in the UK if anyone wished to report or enforce the requirement.

BTW:
PPL (Phonographic Performance Limited) and PRS (Performing Right Society) are the UKs collecting societies

[finite folio]

I have a question about this rights issue. Does this mean that if I have a Youtube, Soundcloud or Spotify account, which are all also public spaces where anyone can listen and watch videos or listen to music, I have to have the same username on Suno? Or what?

[polar spade]

No. Not normally. I can't think of any reason why. Suno is also somewhat a private, members only, kind of program, where, to find your song, I'd almost need to know your user name to search Google.

[woven fable]

Don't known if i can post a link to my song ehre, but i am ahving issues creating a battle duet but the issue is the vocals randomly switch about, even with flawless prompting, and i have no DAW/Extend skills, please help? i have spent over a thousand credits on this SINGLE SONG!

[finite folio]

I have similar problem actually i think.I want to add vocals to an instrumental song that I have previously prepared on my computer with Mixcraft 10 Pro Studio, so that a female alto voice sings a soft "oh-oh-o-oh" and immediately after that a male voice sings a raspy voice "och-och" or something like that. How should I mark this and where so that the AI ​​can understand it?But I still can't add an instrumental to the song, which I wanted. What am I doing wrong.
AI don't understand what I want, I've tried different options several times, but to no avail

[polar spade]

The more you keep your same name to build your brand, though,that can certainly help in fighting or protecting your name or brand. But that doesn't affect your copyright ownership, if there is any copyright to be had.

Other than lyrics, there are few copyrights to be had at the moment. Don't confuse copyrights with just owning a product.

[woven fable]

unless i give the song to someone who knows the editing side of things and they can do it for me somehow?

[finite folio]

If you just create the base of the song with Suno and play everything else yourself or create it in a separate music studio on your computer, I mean the instruments, then 90% of the creation is my own creation. and Suno has only been used to create the base or what about that?

[woven fable]

i am not a musical genius, have no skills other than writing and promting

[finite folio]

Copyright should be clearly understood worldwide, not that every country has its own laws that are different from others, making it so terribly difficult to understand anything at all. I understand that each country wants to protect its authors with its own laws, but for authors who operate around the world, it becomes difficult when there are different laws everywhere.

[polar spade]

I don't like trying to answer here how to do a duet as each person's lyrics, prompts and arrangement as well as composition are different which all affect how Suno perceives what we would like.

One slow but steady learning process is to 1) search "duets" or "make a duet" in this channel as far back as a year.. Read through some threads and cherry pick some techniques to try, understand Suno's thought process and realize how we can't easily force a change in instrument, whether we go from a Heckle Bassoon to a Bass Bassoon to a Contrabassoon, or we try going from lower "male" frequencies to a mid-range blend to higher "female" frequencies.

And : 2) find references by searching "duet" on the Suno Home page and ones that display lyrics and metatags to give some insight to what might have a stronger influence.

Typically, as the tools become more available, I've skipped worrying about my tags working fully on first try and just go to editing and replacing sections to find the suggestion having the gender/instrument I want. Then ggo to the next edit. At the least, if I start edits early I'd be spending nearly the same credits or less and getting some good skills in the process.

[polar spade]

If you wish to try a "trick" : duplicate the verses or chorus within the same composition (Create page) to see if Suno sings differently between them, chose the ones that are correct and remove the others.

Obviously limited to 5000 characters, so we might have to do this in sections.

Extensions are a good and basic skill to have and are our friend when needing to create a fast and pin-pointed song.

[woven fable]

this is the issue... i ahve done all the research on the master prompt etc... its aparently a case of every single line has to be [male]/[female] Block and the style of the music they are using and in the style prompt you state [Female sings] X STRICTLY [Male] Sings Y STRICTLY

and have spend over a thousand credits, and it still switches ONE SINGLE LINE which is cannot do in the thing i am trying to do... 3 other duets worked flawlessly after only three spins so its not a user issue in my opinion

the issue is, i need to keep the single pbroken line

[willow jasper]

Appreciate this, that’s what I seem to be seeing too. I have reached out to Suno support email too for an official response for my own documentation purposes.

[woven fable]

if i could link the song you could see exactly what i am doing

[polar spade]

No link needed. I believe you. Changing out the gender in one line might take running Replace on that line, adding style prompts to force the genre (kpop, beautyshop, opera, femme folk, musical, etc) to get one sample where one is female. These styles are "indicators" of a common thread, the female voice, for instance, to Suno and mostly do not influence the genre, though it might happen.

[woven fable]

thanks, but i need pro advice... (no offense - i mean, i know all that) here is the prompt: "Opening with sardonic, flirty female spoken-word layered over a dark 3/4 synth drone and amp noise, the song erupts into 4/4 with searing Celtic punk fiddle and raw male vocals. Rhythms keep fracturing—gritty 3/4 waltz fills, crushing grunge guitars, and wild drum stutters collide, pitting swaggering 4/4 against lurching 3/4. Female Gothic Waltz sections stake out brooding synth and fiddle textures, while male Celtic punk sections unleash explosive punk energy, all underscored by abrupt stops and chaotic instrumental swerves; vocals never overlap, keeping each section fiercely distinct."

and the single line which is done by the male is
[Female - 3/4 Gothic Waltz] My brother taught you how to [Female - 3/4 Gothic Waltz] hide your scars. I'm going to [Female - 3/4 Gothic Waltz] teach you how to use them.

[polar spade]

No offense taken. Hope a pro is around to help you further than what I could offer. I'll step back so as to not waste your time further. Good luck.

[quiet solar]

If the only songs that you're playing were created by you, and therefore you own the copyright and have a commercial license to reproduce them as you see fit (which you would under Suno Pro) and you're not charging yourself royalties (because you've not gone completely mad), then you're fine. You can broadcast them to the public as much as you see fit.

However, DasVon is absolutely right - in practice, you're going to want to think about how you would easily prove to someone asking that it's your music, because PPL and PRS conduct enforcement very similarly to TV licensing. They'll drop in in stores that don't have a licence and, if you're playing music, will immediately assume you're breaching copyright and start throwing threats around - the onus falls onto you to prove that you're not in breach.

Should be as simple as being able to show the playlist being played, and show that it's a perfect match for songs you've made in your own account in Suno.

[polar spade]

The Berne Covention of.... Emmmm... 1887, I think, made it easier for nations to help each other out in reporting and payment of royalties. For the most part, the rules are about the same, with some countries being more lax and some things being more expensive in other countries. I mean, filing for copyright is much easier in Argentina to then apply to the USA but then it can be $500 a song too. 🥴

I've looked for any loopholes I can find. Lol

Just write Christian and/or Kids song, no one is mean enough to file an infringement against you. (kidding... but true.... but kidding)

[graceful condor]

If you pay for sub with Suno and are not a free user, the rights of what you create belong to you. My friend got over a billion views on one of his Suno singles. He even hired a distributer on Spotify to promote his song. Just to clarify.

[polar spade]

Perfect. Looks like you've covered your assets. The more we know the better. 🙏

[polar spade]

Woop! Woop! Way to go - - your friend! Good to hear the good news. Hope they keep it going.

[quiet solar]

Duet prompting is infamously tricky and unreliable in Suno - there's no way to "command" Suno to do exactly what you want, so from that perspective no, it isn't a user issue but a model issue.

Either you can wait for Suno's model to develop sufficiently to work accurately with your current prompts, or there are ways to tilt the odds in your favour with tight prompting, and with a bit of dedication yes - there are several routes to create a song purely in Suno without resorting to an external audio editing tool. Though if it's complicated, and you're not using explicit workarounds, spending over 1,000 credits is not at all surprising.

I've helped out a number of people with prompting for duets. I've even fixed the offending line for people in Audacity when they were struggling with a single line, didn't know how to use an external editor and came here seeking help.

But if you're going to be that rude to DasVon, who is regularly one of the most helpful people here, and who suggested a workable option that you rejected out of hand, I'm inclined to leave you to your own research and wish you good luck.

[graceful condor]

He went viral international, but his song was nuked off of Spotify and other platforms for political reasons. I just wanted to help the guy asked about ownership.

[woven fable]

it asn't intentional rudness, hence the no offense ... i am autistic by the way... no excuse... but my point was only to say hey, i have learned a lot with suno etc, i need further help which i should ahve said from the startt, but stress + ASD + ADHD is ell, you either get it or you don;''t... no rudeness meant at all

[finite folio]

500$ its pretty expinsive 🙄 😕

[polar spade]

Well, the initial rush was something though.

Shred the gnar ‘til the tube caves, brah. 🤘

[quiet solar]

Alright, DM me a link to the song so I can see the prompts and lyrics, and tell me which bit it's getting wrong, and I'll see what I can do.

[woven fable]

thanks and sorry for any issues @polar spade

[finite folio]

I am also autistic but it is not very serious for me, but I have learned and trained myself to keep myself in check for a very long time. Everything is achievable. I am not saying this in a bad way, but I just wanted to share my personal experience of coping with it 🙂

[finite folio]

Do you know these things Kanthric, I need some help with one of my songs, which is instrumental but I want to add some vocals, both female and male.

[tidal hare]

Hello! I am looking for advice on how i can build a strong Personas with consistent vocals, style and dialect / pronouncing. I always use my own recordings when generating tracks and usually work back and forth with stems and adjusting vocal melody by singing myself or clipping/moving the Suno vocals (thanks to advice from here earlier🔥 ) and then reupload to Suno. However, i have not nailed how i can create a Personas yet that will fit my need. Should i focus purely on the vocals, or should the song genre, style and instruments also be similar to the sound i want? Any advice would be appreciated😁

[finite folio]

How i must put that female voice sing and next say male voice ? like this
[Intro Instrumental]
(1.female alto vocal-"oh-oh-o-oh")
(2. male alto vocal-"och-och")?

[steep haven]

I created a song that turned out really well, but Suno, for some inexplicable reason, replaced the female voice with a male voice, which doesn't match the persona I used to generate the track. Is there any way to use the track—which is actually quite good—by replacing the male vocal line with the voice that's part of the persona? If so, will this require using the Studio functionality? I would appreciate it if someone could guide me through the steps.

[raw oasis]

I feel like these conversations might be coming back up frequently. You might try searching previous conversations on duets using the Search feature here:

[finite folio]

I can't format my query for the search, my English is not that good. 🙁

[raw oasis]

Try just duets and press enter. should give you lots of gold to search through.

[raw oasis]

You can even look for the results labled tech-prompt-talk and when you hover over that result can Jump to that section.

[finite folio]

Ok i did set this that exactly what there Windypier writer
[Chorus Instrumental]
[female alto vocal] oh-oh-o-oh
[male raspy vocal] och-och
for example.
But still there have female vocal but not male.
Question is, why Suno generate there female but not male voice?
I let made Celtic, Nordic folk, metal, world mixed style

[raw oasis]

It is not something nailed down that Suno can do reliably yet.

See how I just changed the lyrics in a test to include words "She" might say... and then words "He" might say?
https://suno.com/s/lTUDUlYJJUFofCjF

[finite folio]

yes why this does not work for me 🙁

[raw oasis]

and also not everyone will want to include pronouns in their lyrics.. so it is tricky.

[finite folio]

https://suno.com/song/40bfda25-5a6a-4e95-b128-53980a24401c female vocal there is but not male vocal.

I did more 3 and all have same. I did cover of track what i did long time ago with music studio what i have.

[raw oasis]

I find duets distracting.. would really need solid personas like Meatloaf's "Paradise by the Dashboard Light".

or Andrew Lloyd Weber's Phantom of the Opera

[bronze harbor]

Billion?? seriously?

[tepid moth]

Okay, so I created a song in Suno. Extracted the stems and worked on it in my DAW, with layering my own sounds, and kept only the vocals from the suno creation. Tried then to reupload it into studio to create additional stems, and got this error message: "Uploaded audio matches existing work of art". Will this say that Suno then generated a copyrighted song in the first place? I dont get it.

[raw oasis]

I have gotten that before. How do we get around it?

cause that is going to be key.

[tepid moth]

is it because I already created that song in Suno, or is it actually that the song it created is copyrighted and used before. That is much darker, so if so I delete the whole song of course.

[finite folio]

There are also super good voices in Celtic and Nordic folk and folk metal, I can't name them right now, but some female soloists are like angels on stage. And men in metal have super nice harsh voices, women also have really great metal voices, for example, Tarja Halonen, the first female soloist of Nightwich now Floor Jansen for example. Or Arch Enemy and Angela Gossow for example and have really super death metal voice.

[half dragon]

I often took my old Suno songs generated from previous model to remake it by Cover it and my final version usually done externally on DAW.
So I tend to upload the final version back to Suno and Cover it, but it did happen to me too.
It happened on one instance so this might not work for you, but try it ...
I renamed the file some random name like... 1.wav and it let me uploaded.

[tepid moth]

I try that now, but since I do not know why this error message occured its kinda scary too move on with it

[raw oasis]

I can understand if it thinks you might be uploading another persons Suno, or other artist.. but the ones we create and extend in DAW would be great to reupload.

[half dragon]

Unless you know you're using someone else sound/music - you're ok to upload your song back to Suno.

[bronze harbor]

Would be helpful if Suno reported back which song it's violating. Hopefully they do this in the future.

then you'd know for sure if it's the song you created earlier

[raw oasis]

this might help a little.

[tepid moth]

I renamed the bounced version from my Daw, and then it worked.

[bronze harbor]

I think you can breathe easy. Suno's algo must be doing some simple name checks to start.

[finite folio]

Sorry but what is this SunoMate?

[wise kindle]

don't even bother

it's not affiliated with suno, so don't let the name trick you

[raw oasis]

If we mixdown from our DAW's should be a unique filename anyway

[finite folio]

ok. thank you

[bronze harbor]

gonna test the Suno App. Ignore me.

@shy yew can I use profanity in my lyrics?

ok it does nothing 😄

[raw oasis]

Okay great - also was able to upload a DAW mixdown of a Suno work i heavily edited. ✅

[bronze harbor]

how are you all dealing with clicks/pops in the stems? RX 11 or something similar?

actually this is wrong chat for that. sorry

[raw oasis]

My first song in Suno with similar voice to mine. 😄

though Suno always next-levels things..

[bronze harbor]

Would you ever use an AI service that mimics your actual voice and use the Suno stem as guide? You could end up with your actual voice but singing at next-level!

[echo moat]

need some elaboration here. Made 99% the perfect song, i want to grab the stems so i can boost the drums, it changes the actual structure of the song, any reason for this?

[raw oasis]

tempting at 55 and cannot sing like I once did hehe.. but probably not is first gut check. I feel like most "listeners" to music want the singer voice more than anything to be authentic. So hunting down singers irl.. lol

[bronze harbor]

Wow I've never had this happen. It can introduce some tempo variance but not change the structure! That's an AI fail.

[echo moat]

it may be ableton warping my track as well, just started using it lol

but i notice when i grab stems that it does different versions and the different instruments are changing

[raw oasis]

close but it's been sunofied. https://suno.com/s/bHg7HpOR4NSSDOS1

[polar spade]

You can stem out the instrumental then edit/ add vocals to the instrumental stem and paste lyrics in to try to get the genders to work right again.

And a sample of the steps:
Stem out instrumental - - - >
Add vocals: paste in lyrics - - - >
Click Create

Extended to shorten. Then cropped the end to make it shorter. No other tools. https://suno.com/s/L3EWPtqQiDd2PmNC

[echo moat]

if you get stems, how do you access them again once you leave that screen? do you have to spend more credits?

[half dragon]

No, definitely not.

[honest heath]

they are in your library, but the filter excludes them by default, you have to turn them back on

[half dragon]

And they show up if you open the song in Studio too.

[frail marlin]

How do u edit reverb, echo, etc. ?

[restive pebble]

@bronze harbor
I hope you are there blaring loudly at my funeral.
Just adore the name!

[half dragon]

Never happened to me (yet) - but did you have those in square bracket [...] assuming you meant the style mentioned in the lyric panel?
But I did hear many has experienced it.

[polar spade]

Truncate any phrase down to one to three word prompts. Test to see if the outputs stay effectively the same, but without the style phrases being sung in lyrics.

I tend to only do single prompting, so if I want a producer tag or phrase randomly sung, I put the phrase in quotes and stick in the style prompts.

Oh. You use single word prompts in the style prompt box and it still sings out the words the song? Or do you use sentences or phrases in the style prompt box.?

Oh sorry... You mean the bracketed style prompts within the lyrics box/panel?

My misunderstanding. I thought you meant style prompt box or panel below the lyric box. Nevermind then.

Can you remove or backspace out that line space in your prompt box? Try for no paragraphs in the style prompt panel. Only commas or if you must, periods/full stops.

If you had copied and pasted in the styles, you may also have other hidden commands needing removed.

".. . . Still happens..."

After removing the line space and separating all with commas?

Also, you have exactly 1000 characters of 1000 limit in the prompt box? Keep in mind, the closer these boxes come to maximum capacity, as well as possibly having a long title often from multiple edits, this can also affect the outputs in unusual ways.

Moderation helps.

Cool... glad you worked it out. It can be just something small like that.

[worthy dragon]

I cnt post in showcase

Okay thank you

But i never posted

And it still wont let me post

I am new here

I may need permissions

Its greyed out

[worthy dragon]

??? This

[half dragon]

Yes, sounds like you need to post direct Suno link to the song.

[worthy dragon]

I got uuu

I did

[neat nexus]

Note the time

[worthy dragon]

I can post one more in six hours

[neat nexus]

Two more. In six hours.

[worthy dragon]

Okay okay ty

[neat nexus]

Two links to be absolutely clear. You can post once every six hours. In that post can be two links.

🚨 Share your Suno music here! All links must direct to a single Suno song, regardless of the platform. Suno profile pages and playlists are also allowed.

You can only post one message per 6 hour (2 links per post max)—so make it count! Don't use other users to double post in those hours, or make it an ad. This is showcase, just post your song, not a wall of text.

[raw oasis]

Was getting this horrible level change/glitch in my DAW after extracting stems and setting up 2 versions of stems for each instrument.. both versions in the Refrain transition sounded horrible. thinking I had to fix it. .. but then i went up to the Suno editor to see if I could rewrite it there.. and low and behold the problem is not even there and transitions nice in Suno. quandry now...

Only usable as produced by Suno if this is the case.. any luck Extracting Stems again?

[knotty stag]

I have tried to use a stem as a persona, but Suno would not let me, or is there a wayt that I missed?

[lapis snow]

Has anyone lately have low audio quality and noise problems in V5?

[knotty stag]

Switch to Linux. It is tricky and time-consuming, but worth it in the long run.

[half dragon]

Should be just as simple as select the stem, then Create > Make Persona
What error did you get when it didn't let you?

[earnest prism]

hi, is anyone had any problems with studio draining credits but not generating anything? I uploaded my own sample and try tu build song around it but it just show an errors over and over

[half dragon]

Did they take the credits when error occurred? I was under impression that they wouldn't.
Also, generations under certain length would refund the credits as well.

[knotty stag]

The option to create Persona just was not there at all.

[earnest prism]

yes, credit taken and error notification occured. I already created a ticked. but i couldn't find it directly on website

[half dragon]

It shows up for me and it has "Pro" next to it.
Since you could Get Stems, I assume you have Pro so that's strange - was it just on this Stem on this song?
What about others in your Library?

[knotty stag]

With original song, Make Persona appears, on the stem for vocals, it is not there, see these small screenshots.

[half dragon]

What about other vocal stems? or just this one
Oh and report it if you haven't #🪲┃bugs-web

[echo moat]

my keyboard stem is plagued with micro stutters, what's the best way to fix that?

[half dragon]

Regen it.

[echo moat]

cover or remaster?

[half dragon]

Just the keyboard stem? do you have access to Studio? regen just that instrument there.
If not, then try Editor to fix that section.

[echo moat]

it's multiple sections, and i never heard it when i made it in suno, but then moved it over to ableton to do some drum maxxing and now i can hear it

i'm not clipping or anything so not sure why

[half dragon]

if it's subtle - maybe Remaster could help.

[fringe shale]

Hi whatws hte best formatting to use in the lyrics to specify voices?

is it (Male voice) or [Male voice]? etc

is there a guide for this

[half dragon]

Should be in square brackets.

[fringe shale]

thanks

any tips for having it alternate between two voices?

like a male and female

[near flume]

Ending support means there will be no further updates to fix bugs or enhancements. If you having browser issues I'd suggest switching to a different browser and keep it updated.

[near flume]

We do not "own" the songs created by Suno. We have Suno's permission to monetize them.

[lyric kestrel]

Hi i'm struggling to get Suno to try match something like this:

https://www.youtube.com/watch?v=5jTG5u0ZSaQ

Any prompts i could use?

[severe kindle]

when you create persona are they by default public or private?

[raw oasis]

Public. I like to turn mine off.

[raw oasis]

what kind of song is it?

[hoary wigeon]

I know people probably ask this question over a dozen times, but is it possible to have multiple personas on a track singing a song at like different points?

[frozen dagger]

it's a bug. If your option to create a Persona is greyed out (unavailable), open the song in editor and Export (Save as New Song). Creates a copy, the copy should be able to create the persona now.

[severe kindle]

negative

[frozen dagger]

have to doa cover using different personas then seperte vocals and edit together

[polar spade]

Yes. Depending on what you mean. We can add a different persona at each segment of multiple extensions, if you wish.

[echo moat]

i'm getting stutter in my stems but not in the actual track, how is that possible?

the synth audio is warping, but i play thew hole track it is not there,

[frozen dagger]

in studio? check orignal/ follow beat settings?

[echo moat]

i don't have studio, that's extra i think

under create, i play my track and it's fine. i go to get stems, see extracted stems, and they all warp

[frozen dagger]

it sounds like what youre hearing is what happens in the procces of the way it is generated and then separated. its not a true stem. it will have weird inconsistencies and artifacts. the track wasn't built from individual stems.

[echo moat]

ah ok gotcha

what's the best way to fix it after adding my DAW changes? i got like rx11 or i could remaster?

[frozen dagger]

if songs were a salad normally, you could pick out the cucumbers you didnt like, but suno put that salad into a blender before it shaped it into something that looked like a salad and now youre trying to diffuse cucumbers out of a artificial thing... that probably made things more confusing.

[quick sonnet]

4.5 is defenetly a great improvement compared to 3.5 in quality, track length and many more, but one single thing 3.5 seems to be better, when it comes to covering an existing audio and want to keep it the same, or at least very close to it.

[unreal prism]

Suno is Excellent for progressive metal & djent 🙂 [see all video]

[graceful condor]

You're right but Mac or Linux is for rich people. I don't have the money for that.

[knotty stag]

I tried that and it worked, thank you.

[reef pendant]

Could you share (either here or via DM) a few examples of songs where part of the styles are being sung as lyrics please? (Drop them into a playlist if it's easier.) In the screenshot you posted it might be caused by the ( and ) which are typically used for backing vocals, though having it split across multiple lines (as Das Von mentioned) may also cause this. Essentially if a [tag] or the style description look like lyrics, there's a chance that it can end up being sung.

[graceful condor]

The frequencies from the sounds are pulled apart stick with the other frequencies in stems. There are better stem splitters than Suno out there but you have pay for it. Maybe take your stem, upload it and make a cover of it. I haven't tried that yet. Remaster and remix never sounds good at all when I try it.

[frozen dagger]

I don't have experience with cleaning up the audio or mastering. I mostly experiment within Suno. So, if I was working within Studio trying to clean up a vocal stem's audio. I'd be trying to take it into editor, clean up what I could with cropping. Copy and Paste the parts where the singing is similar to sound to what I would want in the composition and place as I want. Don't have to really worry about lyrics, just want ambient noise for tone and texture etc.
Each iteration is going to step away from the original and experimentation would be required. Some good, some bad. Using this method I used a persona and I'm pretty satisfied with the consistency over multiple (hundreds of) gens.
If I had the knowledge/skill I wouldn't be foolin' around playing with "what can I do just in Suno". Best/expert results are going to be from combining tools and using human and machine in harmony.

[graceful condor]

Yes, download the ones you like, slice them up in DAW, reupload to Suno and create a persona of it. I think you're asking about creating a mix of different sections from multiple personas, like the image here. I have 30 of the same song mixed together as one track.

[echo moat]

is demuxing decent for stem splitting?

[void turtle]

What's persona ?

[misty girder]

Hello everyone. I'm a multi-instrumentalist songwriter (corporate worship, country, pop, and rock) and I'd love to be able to have Suno create "demos" that are high-quality without having to spend hours tracking and/or hiring musicians. I can track roughs in Logic Pro. What’s the best way to get “fully produced” versions of uploaded audio? I’m tracking acoustic, bass, drums, electric guitar, piano, and vocals but the output of Suno is “wonky”

[restive pebble]

A persona is a type of Style captured by Suno for use later in a different song. Usually used for vocal consistency.
To make one:

1. click the three dots to the right of a song.
2. hover over the Create option in the menu.
3. Click Persona if available.
4. Name and describe, then save.

you can then apply it to a song by choosing it from on of the three options above the lyrics section. +Persona +Audio and +Inspiration

[void turtle]

Thanks !

[quiet solar]

Wouldn't we all... xD

When you say it's "wonky", do you mean that the output needs more detailed mastering work to bring it to professional demo quality, or that the sound itself is poor quality? Or...something else?

[quiet solar]

If you're a Pro/Premier user, you own the songs to all legal intents and purposes. From Suno's own TOS:

Subject to your compliance with these Terms of Service , if you are a user who has subscribed to the Pro or Premier paid tier of the Service, Suno hereby assigns to you all of its right, title and interest in and to any Output owned by Suno and generated from Submissions made by you through the Service during the term of your paid-tier subscription. However, due to the nature of machine learning, Suno makes no representation or warranty to you that any copyright will vest in any Output.

The previous section of the TOS essentially requires you to grant permanent, total and irrevocable rights to use the music freely as they see fit without any claim for royalties, in exchange for using the service. But the ownership any any copyrightable content itself vests in the creator.

[misty girder]

Well, it’s probably related to my prompting tbh

I’m sending it, in audio, a solid draft of what the song is, the liberty it takes aren’t really “on point” with what I’m trying to get from it

It’s not a quality issue at all
It’s always just a “why would you think that’s what I wanted there?” sentiment….specifically when I’m sending it audio that should confirm what I’m trying to accomplish, to some extent

[misty girder]

It’s beginning to make me think that I’ll need to run multiple “add instrumental” attempts against my uploaded audio and then export the stems, bring them into my DAW, and then keep what I want/discard what I don’t

Am I right to think that “add instrumental” is what I should be doing (with uploaded audio) to accomplish my goals?

[quiet solar]

Ah, yeah. I feel your pain. Sometimes working with Suno is like sitting down with the world's most ornery session musician.

Starting with the basics, if you're coming in with pre-existing audio that's already in somewhat reasonable shape, what might help as a first pass is to do a Remaster of the audio upload with audio variation set to subtle.

If that doesn't work (it will depend a lot on genre), then doing a Create with the uploaded audio added and the audio influence set very high (e.g. 85%+) will help. For extra emphasis, you can create a persona based on the uploaded track and add that in too.

Then if you're comfortable with style prompting, then you can prompt a description as close as possible for all of the key instruments and vocals in the style and put the style influence slider up as well - if you're not, keep it simple and keep the style influence slider low (e.g. <25%). Turn weirdness right down to <15% - you don't want Suno being too innovative with your track.

If you're past that point and it's still consistently messing things up, then yes - using Remaster, a new audio-uploaded creation or 'add instrumental' for a specific stem via Studio, and then slicing out the stems or bits of stems from what it creates that you like and taking it them to a DAW for further work is a pretty solid suggestion.

I've not used the 'add instrumental' feature extensively, so I can't really comment on whether that's a better route or not.

Also, as a general rule, avoid setting the sliders too close to 0% or 100%. Tempting though it might seem, Suno tends to do weird things at the very margins - and it's usually at its best when it has a bit more freedom to work as designed.

[misty girder]

I’m gonna give this a try!

I’m basically trying to give it “roughs” of all of the tracks and get it to “sharpen them up to pro quality” and have a little creative liberty with the production

To be clear, I’m feeding it “roughs”
Imagine it’s a vocal and acoustic demo (though I’m sending more instruments) and on acoustic I’m strumming heavier and keeping rhythmic timing etc so any HUMAN would be able to decipher “oh it’s a build here” etc

My goal is to have it take my “demo tracks” and make them “studio quality” tracks while also giving it a bit of liberty for a drum lick here or a guitar run there etc

[mossy raptor]

Hey guys,
I tried looking via the search function, but i couldnt find anything that I haven't tried already.

I am trying to make a song with both male and female vocals, that sind certain verses, and sometimes both at the same time.
Getting two sets of vocals into the song in general works like a charm.
However, they keep singing parts in the wrong voice.

I tried like so:

[Verse 1][Male vocal]XXX
XXX

[Verse 2][Female Vocal]YYY
YYY
YYY

[Chorus][Both]
ZZZ
ZZZ
ZZZ

Anybody got something like that to work well?

[manic vapor]

[Verse 2; Female]

or sometimes extending where you need to with the headers work too

[quiet solar]

I've had mixed experiences doing something very similar. Sometimes it feels like a struggle to get it to stick to the brief. Other times I'm quietly impressed by how it perfectly replicates the strumming and subtle flourishes in the vocals.

And on a couple of occasions, I've been absolutely blown away with what it's been able to build based on a rough acoustic sketch. v5 is certainly showing some really, REALLY good potential on that score.

Best of luck to you!

[mossy raptor]

Oh I haven't tried with a semi-colon yet. Ill give it a shot.
Most often it messes up after the chorus, because I am following an unusual pattern of:

Verse 1 Male
Verse 2 Female
Chorus Both
Verse 3 Male
Verse 4 Female

Suno nearly always switched the voices for 3 and 4 around

[hoary wigeon]

Thank you! 🙏

[quiet solar]

This question comes up about 4-5 times every day.

It's very difficult to get Suno to reliably switch vocals at the correct moment. There are things you can do to make it more likely such as specifying two distinct vocalists in the style prompts and keep a clear structure for how the singers switch between each other. However, how the genre usually handles two singers (especially male/female) will often have a large influence on the output.

Your basic approach is fine. Just expect it could to take a LOT of regenerations if you're trying to get the song correct all in one take. So you may want to consider approaches like:

1. Extending from versions that get the first part correct and gradually building up the song that way, banking the bits that work as you go
2. Doing an initial version with only a male singer, then using Extend to create a second copy of the song with only a female singer, then extracting the stems and using Suno Studio or an external editor to patch the correct singers into the correct places against the instrumental track

PS. [Verse - Female vocal] and [Verse; Female] should be functionally identical in terms of how Suno interprets it. Neither will guarantee that it will listen to your style prompt and interpret it correctly at the crucial moment.

[mossy raptor]

Thanks a lot! I guess ill have to look into how the extend feature works. I have not really dabbled with Studio or Editor too much yet

[manic vapor]

use the extend from the create if you arent familiar with it

[quiet solar]

Extend creates a new song using your previous one as a prompt (there are a few ways to do different variations on this theme in Suno). You can combine them afterwards into a single song (or multiple extend paths into a single song at the end).

Suggestion #2 really requires using either Suno Studio specifically or an external audio editor. The basic Suno internal editor isn't great for doing that kind of work with stems.

[graceful condor]

Try adding 'sing the vocals with precise accuracy' in your prompt as well as adding more gender description in your vocals section where you put your lyrics. You can try duet or duel too. It shouldn't take very much effort with properly defined language for the Ai to communicate with. It will happen eventually. You can try external tools outside of Suno if you really want to dig your heels in. Example, I guess. https://suno.com/s/OukH3vqpfmL9KO1s

[graceful condor]

Sorry, I put the wrong song, you can check this prompt, with adding the 'precise accuracy'. https://suno.com/song/32d1e651-dccc-4edf-b9a8-56c092e4d2ec

[polar spade]

Are you pitching to artists or what's your goal with the demos?

[near flume]

they're assigning rights to it, no where does it say they turn ownership over to you.

[near flume]

works every time in v4.5/5 for me, but I use Vocals instead of Vocal in the metatag. Don't know if that makes a difference but I've been doing duets since v3.5 so maybe Suno understands what I'm doing from past generations.

[polar spade]

I did this today for just showing how to add vocals to an instrumental, so I kept it short, but made it bounce between gender(mostly from boredom) just to say, it's not, like. . . impossible. https://suno.com/s/L3EWPtqQiDd2PmNC

Regardless of the male/female prompting, the one wrench Suno tries to do in most every output is that it's as if Suno divides sections up like splitting a deck of cards then shuffles the prompts and the expressive bits throughout the song.

Spacing is helpful for changing direction in genders, but then spacing also signals Suno to "fill in" with bits and pieces from the present prompts or ANY bits from any previous edits.

It's always done this since last year. So that will tend to affect the patterns expected of which gender is to sing next.

That's to say, there's no one way or trick to get duets to work, since Suno throws in variants to shift where you my not want it. Best to just throw what you know against the wall and see if it sticks - - if we're going to try to do it in one run. Then edit the heck out of it.

[polar spade]

Interesting. I'm the opposite. I use "vocal" instead of "vocals". For me it makes sense to infer only one person is singing than an ensemble. Each their own. Probably works fine either way. 😅

[restive pebble]

For me what works more often than not i to use granular tags in the lyrics field along with my specification in the style field. It ain't perfect and the Ai doesn't follow the prompting exactly. . .
Style section Specify Call and Response Duet or Duet Call and Response.

Format the lyrics

[Verse 1]
[Vocals: Male, Call]

[patent condor]

Does anyone know if there's a way to use remix to add additional instrumentation to a song in a spot where there was previously none? I want to add a background of flutes to a specific section of a song where there is currently only vocals, but so far all I've been able to do is to change the instruments used in the parts of the song that already have instrumentation.

[half dragon]

I think you can do this in Studio - if you have access to it.

[quiet solar]

Again, from Suno directly:

"If you were subscribed with a Pro or Premier plan when the song was created, you are considered the owner of the song. You also retain the rights to commercial use for the song, even if you end your subscription."

Suno retains ownership of songs created on the Basic (free) plan, but allows you licence for non-commercial re-use of the music.

[wise kindle]

having A LOT of trouble reworking in the instruments in a song. I love the overall melody, so I'd like to keep that the same. Is the best way to change the instruments to cover the song, remove the entire prompt, and replace it with how I'd like the instruments to behave. Or am I supposed to keep the entire prompt in there and just change things around? I've been trying that last method and it's 100% not working

it's just small tweaks I'm after, like a more forward piano rather than subtle felt, and a driving pulsing bass rather than a quite sustained sub bass

[autumn echo]

Hello everyone

Got one technical issue to that need help

The Suno on my browser don't have the Visibility option to change my song from private to public

It only shows Allow Comments, Allow Remixes and Pin to Profile

[wise kindle]

I don't think there's any true private option anymore. By default they are link only (meaning you can share the link and people can still reach it, but they won't be able to organically find it). When you publish it then becomes public

[autumn echo]

Oh, but when I use the search option I can't even find my songs.

[wise kindle]

were they published?

[autumn echo]

YEs

[wise kindle]

If they were published then it should be present in searches (just tested a few of mine and found them). Suno has A LOT of tracks, so it can be hard to spot it at times

[autumn echo]

Oh ok then thanks 🙏

[wise kindle]

just make sure they look like this

[autumn echo]

Yup

[wise kindle]

if you drop the title of the track I can see if I can find it on my end

[autumn echo]

Cat talk by Topdeot

Thanks alot 🙏 🙏

This the cover

[wise kindle]

the search feature is pretty bad. I'm getting completely irrelevant titles lol

[autumn echo]

Yeah hahahaha

[wise kindle]

the more important thing is that published songs are eligible to be placed on the home page for staff picks, trending, etc.

[autumn echo]

Cool

[autumn echo]

How can I find your works

[wise kindle]

I won't advertise in here, but my suno name is Cadenza Vale

most of the time I cover original uploaded songs

[autumn echo]

Oh cool, I'll check it out

[torn spoke]

https://huggingface.co/spaces/TheStinger/UVR5_UI

bruh

It's a gif

woodhouse

[torn spoke]

https://tenor.com/view/rpx-syria-mic-drop-gif-19149907

leaderboards for all the different models too,

[wise kindle]

this server has broken me. I read "Woodwinds" as "Woodhouse"

[torn spoke]

just tested the de-reverb one, actually worked so well

insane, lemme get a before and after

the one with a short name is before, and the long name is after

[torn spoke]

Vocal separation comparison along with a link to the page, and a screenshot of the settings I used.
link: https://huggingface.co/spaces/TheStinger/UVR5_UI

#📚┃suno-school message

that's pretty damn good

[mortal harness]

Can you actually get decent vocal stems with Studio somehow? All the stems I try to get for my songs just regular are all garbage with massive goofy lisps on everyone, so I stopped wasting credits.

[torn spoke]

#📚┃suno-school message

try that service with the settings on the screenshot

[torn spoke]

#📚┃suno-school message link

to the page

[mortal harness]

Oh, I thought that was studio for some reason

[torn spoke]

studio on the left, webpage on the right

[compact helm]

Good Morning,

I tested it once.
It seems to add weight, as it's as targeted as possible to get the desired instruments. You can probably control the intro and outro in the same way. It's important to note that this should also be stated in the lyrics, and this in turn should be stated in the prompt...

for example:
[slowly intro]

[Verse 1]
In the shadows of his soul, lies an angel...

[break down]

[slow pre-chorus]
In the darkness, he tries to hide...

[uplift]

[Chorus-powerful]
Broken wings, (shattered self-confidence!!!)

[instrumental interlude]

[break down]

[Slow Verse 2]
Bloody tears, on the dusty ground.

[Slow Dramatic Bridge]
He longs to fly, to break free.

[Intrumental Interlude]

[Uplift]

[Chorus-Powerful]
In the darkness, he tries to hide.

[Slowly Outro]

Prompt:
Cinematic metal: 0.8, Classic: 0.8, Piano: 1.0, Strings: 1.0, Intro Classic: 0.8, Chorus Cinematic: 0.8, Bridge and instrumentals Piano: 0.5, (bass drum, female vocals, atmospheric, sad, dramatic)

Please note the brackets are for clarification only... the parameters in the brackets are fixed, and the model seems to arbitrarily mix them into the parameters mentioned above, which I'm trying to enforce.... As you can see, I'm saying that a piano and violins are a must and can't be missing. I ran this test 10 times with a result of 80% that the songs are almost identical in sound and bars,

Sorry For my English

[mortal harness]

Not perfect, but far better than everyone having massive lisps. Thanks!

[torn spoke]

[torn spoke]

You can experiment with the settings

there are so many tools there

[mortal harness]

Yeah, I did until I ran out of CPU, the various advanced settings I tried moving up and down, and they only added the lisp back in, the settings you posted were the only ones worth downloading

[torn spoke]

denoising, compression, instrumental stems

[mortal harness]

Yeah, I think I'll try in some of my own tools a well, but I din't know if UVR will be any better

Suno has always been really rough tryign to get anything out of it with any UVR settings I've tried.

[torn spoke]

asked gpt to help me tune the settings to make the vocals sound like they were never recorded with instrumentals to begin with

[torn spoke]

yeah, hugging face is generally a "pay as you go" service

having some gpu time is pretty cheap

[mortal harness]

It definitely processes things a lot faster than UVR that's for sure, heh.

[raw oasis]

what is an HF space (hugging-face?)

[velvet scarab]

There is also a desktop version, wonder how it compares to web one 🤔

[weary heath]

i have suno studio. but how i got my own live recorded vocals into my song. there is no edit to get autotune or what ever. i dont want to have ki vocals

[raw oasis]

are you able to write the song you want - and then download the stems as Instruments + Vocals - move to your DAW and sing it? That is one way. I am pretty sure even if you upload your vocal track to Suno - it will only interpret it and not be your exact voice. I tried yesterday.

[weary heath]

I have FL Studio but my Problem
i cant remix and master. This is why i want to use Suno

[raw oasis]

I hear you there. I am deep diving to be a Producer now.

fiddled for decades.

[weary heath]

Nice. My Producer is going away. So my last Song was January 2025.. This is why I wait to get my own Persona of my own Voice. I hope it will come 😔

[raw oasis]

I imagine Studio will very shortly (months in AI time) have these abilities more honed in. I think now it is in Beta release.

[weary heath]

Yeah i know. I know there are many problems because Vocal Cloning of real Persons. But I only want to Clone my Voice like Kits.Ki but Kits dont make it so perfect mix and master like Suno Ki

And sorry for my bad english 🙈

[raw oasis]

I would download the stems, sing my voice, upload my single vocal track back up and mix it into Studio.
Hopefully another Studio user can help you soon.

From the Suno-Studio page:

They say Export and continue editing in Your DAW - but at the same time they want to BE the DAW.

[weary heath]

Yes i know. The Studio is very good but i‘m so bad at mix and mastery in fl studio so i will wait. i hope i‘m not the only one that want that suno will bring out vocal cloning 🤗

[sly rock]

Hi guys, I have a use case that I'm trying to figure out being new to Suno. I have a drum loop pattern that I'd like to use as a guide. I want Suno to for example only add a guitar rhythm to it but don't change the drum, and not add anything else. Currently I've tried extend, remixing...but all I'm getting is a full song that isn't following my prompt. Can Suno be used to only add specific elements like guitar, bass..etc and keep the current track as it is? Can it be used as a "session player"?

Something akin to Logic's session players, that is.

[reef tree]

is there a complete FX list created from the Suno side rather than the user side? I believe a few more FX sounds need to be added to the library.

[restive pebble]

I have never found one but someone may know..

[graceful condor]

7 minutes 59 seconds of complete silence?!

[restive pebble]

a new record?
Damn.. never had that happen

[willow jasper]

That’s great, thanks

I plan to keep a Google Sheet of all songs with links to their Suno Player so I’ll use that for proof too 👍

[restive pebble]

had 20 sec of nothing usable

[graceful condor]

Maybe I won something? 🎁

lol

[restive pebble]

silent peace of mind before you lose t all lol

but only 7 m 59s of it

Wish suno would adhere to time or bars so the songs wouldn't run so damn long... SO much editing needed to shorten the darn things

[valid ferry]

Good morning! In my opinion, Studio Suno should be part of the PRO plan. Track editing is the minimum a professional needs to publish music, not an award with the PREMIUM plan. This option should be reviewed by the company's administrators.

[restive pebble]

Anyone know how to get a line in a verse accentuated with more emotion?
I've tried adding more style tags to the line but it isn't parsed.
Ex:
[Verse]
Lyrics
Lyrics
Lyrics [Vocals: Powerful, soaring]
Lyrics

[quiet solar]

There's no guarantee Suno will be influenced by any specific lyric prompt; it'll often strike a balance between your prompts and what's typical of the genre.

If you want to try alternative ways to influence it, the [Passionate] tag can be helpful. You could also try putting the lyrics in all capitals, or in quotation marks "Lyrics go here" as Suno it pretty good at reading punctuation cues within the lyrics box and understanding you're calling for different types of emphasis, or trying to hand the line to the backing vocalists, etc.

[restive pebble]

I appreciate the reply Kanthric
Used caps. Yells. lol.
Will try some quotes and other punctuation for effect.
I know there's no dedicated method to get suno to anything except generate sounds.
I've done the emphasis using many descriptors but suno likes her own styles, but we know this.

it's a blues day and I need that specific vocal on that specific line or in my mind, the song mmisses completely.

Was just hoping someone had a way that worked more than 50% of the time.

[last bay]

Maybe this could work? Haven't done myself #💬┃general-chat message

[restive pebble]

Appreciate that. I'll try just before the word instead

[pale steeple]

Hello everyone, I'm having a huge problem with the prompts. I explicitly tell the editor where a female and a male voice should go. But Suno ALWAYS messes it up. Does anyone have any ideas how to fix this?

[quick sonnet]

Is 3.5 deactivated?

[pale steeple]

Created in 5

Can you activate and deactivate 3.5 separately?

[quick sonnet]

It is not visible any more.

[pale steeple]

But I could still select that. However, only version 5 is active, and that's what the song is output with. However, it simply doesn't follow the specified structure.

[torn spoke]

You can always edit in a different program and then upload to suno again

[torn spoke]

Suno is not good at duets, and I wouldn't recomend trying to do them for a while if you're new c:

It's not officially supported, though it can be brute forced into submission. But that takes time, patience, and credits.

[quick sonnet]

I can't see 3.5 any more, just this:

[torn spoke]

#🚀┃updates message

[pale steeple]

I also explicitly tell Suno that he should only sing a Female voice for verse 1, for example, with the note [Verse 1][Vocalist: solo Female].

After about two sentences, he switches to a male voice.

[quick sonnet]

So now 4.5 is avaible, but 3.5 is not avaible any more, but some days ago it was still avaible.

[pale steeple]

Let's put it this way. This is the first time I've tried working with Suno in a duet. But I'm not very good at it.

[raw oasis]

It was answered already.

Suno is not good at it, esspecially for new users.

[graceful condor]

Here is my Intro but I sure a phuk don't see it in this BETA Editor now matter if I align the lyrics it's not there: [Intro]Tenebrae vocant (the shadows call)
Amor perditus (love is lost)
Spiritus manent (the spirits remain)
Silentium aeternum (eternal silence)

[graceful condor]

When I tried to add lyrics to the intro that's not shown.

[polar spade]

Exclamation points can help as well.

[restive pebble]

thx

[polar spade]

When you align lyrics, do you then type in the intro before the rest of the body and click save? Then open align and re-align, hopefully with it showing the new lyrics? It's been awhile since I've added a lyric intro, but that's what I recall.

By the way, did you get the alignment button to work? It appears so. If you get again in Edit UI, I've found changing the BPM to "Follow Track" will unlock the align option, if that helps.

[graceful condor]

no I still have the 🚫

[graceful condor]

I can hear part of the lyrics. the last two lines of the Intro but the editor is not showing them at all.

[polar spade]

It may take creating/adding a section into the front of the output first. I used to do that to have a blank section, then I could add in what I wished.

Have you added lyrics intros before? Again, I'd have to do it again, before knowing for sure. I'm not in front of my computer at the moment.

[graceful condor]

Yes

[polar spade]

tried this?

[restive pebble]

I got the vocals to accentuate using [Vocal Style: Rising, Soaring, Powerful]
Still not consistent but we all knew it wouldn't be

so maybe it wasn't me, but suno saying let's give this gu as break he's spent thousands of credits on this one song 🤣

[polar spade]

But you promised! 😥

[restive pebble]

Ai take liberties with the creative vision I have. That isn't fair but I think it's great when Ai creates a sound I am impressed by, so I guess I'll take what I can get

Without Suno I would just be a poet. With Suno I play the role of a lyricist, DJ Producer, Artist and groupie all from the comfort of my home. WHo could ask for more

[torn spoke]

Getting a good voice with suno is the exception ngl, ye just gotta hold on to that thing for dear life and mold it with edits and persona's

[graceful condor]

Okay That was it but the section I replace is like 5x louder than the rest of the song

[polar spade]

We can reduce volume per each section. Try that.

[pastel plaza]

hi guys, help plz im lost, why do i keep geting this ssssss sound in the backround ?
https://suno.com/s/VDmnzDLjdX3o64Pj

[graceful condor]

@polar spade Thanks for the help.

[polar spade]

Is it specific to a section of this song or in the whole song? Does it happen in every generation of this song? In every generation of ANY song?

[pastel plaza]

the all long song only in this song

[polar spade]

The best way to know if it's just this song is to do a few more. But generally, a hissss can be:

° just a glitch
° Suno failing to process the overtones very well
° Suno having trouble with certain combinations of genres.

Little tricks like removing a prompts, or adding a prompt could change things.
°. Try, adding section at a time, by using extensions, so Suno doesn't get "tired" in extrapolating each next chord in its composing the output.
°. Run a remaster or Cover to see if we can change the hiss to a synth, a cymbal, wind FX, or other instrument.

Otherwise, 4.5 and 5.0 have been great at NOT having too many or any hissing or shimmer. So if we get it now, it's just cause it's a busy server time of the day, or an update is going on, or any of those small glitchy things that can happen with one song.

Hope something helps.

[ripe galleon]

is it just me or does suno not pick up on the vocals of the input song i give it anymore when i extend it?

[polar spade]

Since it's pretty faint as well as this is an up front vocal, we can't forget Suno is pulling off of training on recorded audio. Folk, acoustic, and single instrument and vocal songs often pick up ambient room noises, mic bumps, p's popping, feedback, audiences, coughing, applause and other noises in the actual recording of the record. The older songs, obviously, there's more of this. So we can actually get those sounds mixed in as well as Suno attempts to keep it sounding authentic, no less than it may add reverb.

Remaster tends to take those out, if they are faint.

[polar spade]

Each extend can vary. Some may sing a lot of the vocals, another might be more instrumental. It's done that consistently since extensions began.

[ripe galleon]

the thing is

im giving it a completely normal song and it just makes female vocals for some reason???

none of the extend geneartions ive done have any of my own voiced vocals in them

just this weird female voice

v3.5 was more consistent with this surprisingly

i dont know if i just had a bad generation everytime or something

[polar spade]

Though it doesn't clone, my experience has been it always picks up characteristics. So yea. A female squeaky voice or whatever isn't usual unless your prompts are steering it that way, such as anime or vocoloid.

[ripe galleon]

it doesnt clone your voice anymore?

[polar spade]

Not officially, no.

[ripe galleon]

aw

[sonic harbor]

anyone know where song upload is now? ive been trying to find it for the past 15 minutes and none of the ais said it was removed after searching

[hard beacon]

create +audio ?? 🙂

[ripe galleon]

should be when you click custom in the upper left and press + Audio

unless you mean something else

[sonic harbor]

oh i didnt realize those were clickable thanks

new ui is so confusing lol

[polar spade]

It's also over here in the Library if you ever need it.

[sonic harbor]

maybe im just getting worse at ui because of ai, i was feeding it screenshots and stuff and the ai didnt know either. mosto f the time thats how i find ui stuff now

yeah i dont have any upload audio button, just in create when i actually try to make a song

[hard beacon]

40 open tasks ????? 😄

[sonic harbor]

ai wouldve seen upload audio but just audio it didnt associate with it pretty much

[polar spade]

You dont? 😅

[hard beacon]

öhm... no ??!!!

😄

sorry, but i am out at this point 🙂

[graceful condor]

'Atmospheric' is going to add a windy-hissy sound. When I prompt for Atmospheric Death Metal it adds a strong hiss sound. I don't like it but that word does set a mood for the song that I like. Unfortunately that awful hiss sound is there.

[hard beacon]

ufff..

you are right.. that's really weird

in all outputs ??

and: what model???

@graceful condor

[reef pendant]

Can you post a link to the song that was completely silent, if you still have it? (If you deleted it, it should still be in the trash.)

[hard beacon]

Just another short "offtopic" question... my credit renewal is in round about 24 hours.. if i buy 500 credits now - will they be gone tomorrow or stacked up?

By the way: I've noticed that for the past few days there has been a slight noise in the high frequency range. It's particularly noticeable in metal songs (hi-hats, ride).

[wise kindle]

only subscription credits expire

(your top-up credits will remain untouched when your subscription renews)

[hard beacon]

ok... so if i use 100 of 500 credits today i have 10.400 tomorrow?

(if i buy 500 now)

[wise kindle]

That sounds correct (but don't quote me on that). If you need some more convincing, this is on the pricing page

Credits included in subscriptions do not carry over from day to day or month to month. Purchased top up credits do not expire, but require an active subscription to use.

[hard beacon]

dumb question coz you give me the answer.. just wanna make sure i understand it correctly 😉

ok... thanks buddy..

[wise kindle]

no prob! so you can feel safe your top up credits won't expire tomorrow xD

[hard beacon]

we will see 😉

[polar spade]

Yes. If you're Premium. 😁

Also, your extra credits should remain and in limbo if you should drop to Basic for any reason. Then active upon upgrade back to a paid account.

[misty girder]

Hey there. I’m pitching to artists mostly 🙂

[polar spade]

Early, on we could double-cancel or basically cancel out our credits if we cancel twice from Basic. I don't think thats the case now, but suggest don't ever cancel while in Basic tier.

[polar spade]

You'll lose a few artists who may approach you to simply use your stems if they don't want any trace of AI in them.

You can also get soft-balled by any publisher as they still never accept any AI traceable tracks - - especially for sync.

Pitchers will also listen to anything, but may never tell you if they simply didn't pitch the song. "it wasn't what they were looking for", will be their go-to reason and not say that it sounded too A. I. to pitch.

Otherwise, as long as artists or managers feel your song can be re-produced and they file their version for copyright, you can get some holds, even with it tooled with some AI.

As for your original question. Certainly the Suno song by itself is pitch ready. But edits are needed and the downloads are wonky, then download and use a third party splitter of course. The more you can diversify and dilute A. I. as a major composer, the better others will feel to use the material as well - - and easier to copyright later.

Don't forget, skipping the tracks and full production and having a crisp home recorded acoustic version, especially for worship, pop and rock, can be good in a few cases. So do the full production with Suno and work it out on just a guitar occassionally is the way to go, but only now and then these days. Only maybe Christian worship... The others lean more now to single artists who will gladly just use the radio ready master tracks you made.

[grave glacier]

What happened to v5 now I get too fake sounds

[polar spade]

It's A. I. It's all not real. 😁

So, you've used Suno for a year now? . What do you mean. Did you get some gibberish in the vocals?

[grave glacier]

If you know how to use you can get realistic sounds .)

Idk why but I started get my old quality finally after 25th try

[frozen dagger]

Sometimes it's like that. I was trying to add a [Laughing Maniacally] in my mix, which it has been ignoring when usually its kinda hit or miss, i thought i guess dont work in v5 but then it spat out a laugh at me after about a dozen gens.

[torn spoke]

did you see my hugging face thingamabob

[frozen dagger]

no, what we talking about?

[torn spoke]

It was a part of my journey to try and figure out if I could run the vocal fixer locally

I figured out that it's not good for anything high fidelity

and instead looked around for ways to simply get better stems than suno does

lemme find a link to the msg

#📚┃suno-school message

[frozen dagger]

i did ctrl + f and looking at your messages now. checking out links

im trying to listen and wondering what the heck is all this noise im hearing. i pause it and still hear it.... had my game running in background....ugh. im tired.

[torn spoke]

HAH

[frozen dagger]

so stinger on huggin face for stem. work on more than just vocals? that reverb removal was pretty darn good.

[torn spoke]

they've got leaderboards for kinds of tools

[frozen dagger]

like really good. that last 3 seconds almost elimintated it

[torn spoke]

all in the same space

you can see which ranking they have on what genre

[frozen dagger]

bookmarked. something else to banf our heads against the wall for

now im tryign to think of something to test it on... what im working on now is just gamer slop lol

[torn spoke]

gamer slop you say

[frozen dagger]

ya. just throwing stuff together for short clips.

[torn spoke]

like https://cdn.discordapp.com/attachments/1431756227427172454/1432808553642590271/shoobadaboo.mp4?ex=69026673&is=690114f3&hm=16f19934b82f8b482672699a4a9cfc14b36fddd97c355aea9b9b7ae6320de5fc&

[frozen dagger]

playing Marvel Rivals with my buddy, adlibbed "Line up the bots, line up the bots. Take 'em out one shot." Now we ar in Suno messign with that and my head is like.... oooh there's something can do with this with the rivals replay and editor

just picked one and the seperation is already something i would much prefer to use

wow. that should save people a lot of time. this is a 7min + song with lots of stuff going on and i only hear the vocals. i wouldnt have to clean this up like would with Suno seperation

not perfect... but much imporved

i would use this to separate a vocal stem if i wanted to use it as a persona because the vocal is isolatd so much better.

[torn spoke]

Line up the bots
Line up the bots
I've got no doubt
Need just one shot
To take them out

idk

couldn't not

[frozen dagger]

haha nice. i bridged..
[Bridge- Aleatoric Call and Response Glitchy Main Vocal and Down-Pitch Vocoded Backing Vocals]
0-1-0-1— 1-1-0-1-0

Won one, now my team's bots.
When I win, 'cus I carried everyone.
Take them down, one by one.

[Dramatic Countdown]
6-5-4-3-2-1—
[pause]
...(none.)

E-O-M-M
"mmm"... tasty
Last 6 not won— my turn now to have some fun.

[Maniacal Laughing Sound Effects]

[torn spoke]

[frozen dagger]

something like "not a second thought" lol

[torn spoke]

Man the suno discord needs a lyric chat

it's funny, that's like, literally the complete opposite of what I want Suno to produce

[frozen dagger]

lmfao

[torn spoke]

the vocals that I go for are the ones in the example clips, as human as I can get em

[frozen dagger]

ya i think that to myself when i read people seeking help for clarity and fidelity. Ask me how to mess them up! lol

in this instance going for robotic/glitch so leaning in. surely will be some Sinatra moments ill want the clean singing on. but yeah. just for fun.

[torn spoke]

I suppose robotic/glitch vocals is what suno would be best at

just by virtue of what suno is

[frozen dagger]

Asked chaGPT: what would be the technique called of fixing an error by leaning into it? (pretty much my style, reflected in purposeful misspelling of my name)

That technique is often called “error reframing” or more creatively, “embracing the glitch.”
But depending on context, there are a few domain-specific names for it:

Context Term or Phrase Description
Art / Music / Design Glitch Aesthetic or Wabi-sabi Turning imperfections into intentional design features.

Wabi-Sabi. That is the approach I would reccommend people use for Suno. Especially newcomers.

[paper dawn]

Anyone know if they’re gonna have a studio competition anytime soon to wear studios the tool you gotta use for the remix?

I don’t know why I tagged you man sorry

[finite folio]

When does Suno ask for confirmation that my previously made mixes are my creation? I uploaded a song I made with FL Studio about 10 years ago and now, Suno asked for confirmation, this song has been uploaded to My Space and Soundcloud for a long time ago.

[torn spoke]

https://tenor.com/view/cat-gun-wizard-wizard-cat-cat-gun-gun-cat-wizard-gif-821142718141522645

[torn spoke]

Suno is asking to protect itself from lawsuits.

... well, the ones they aren't already in the middle of.

[finite folio]

I think Suno can also ask for confirmation if any parts of this song have been requested for use by a well-known DJ?

[torn spoke]

What....?

[finite folio]

sorry my english is not so good 🙁

[torn spoke]

Sorry I don't mean to be rude

[finite folio]

This song, which I made in FL Studio about 10 years ago, I uploaded to My Space, where a well-known DJ asked me for permission to use some parts of my music in his mixes, and I gave him permission to use them.

[torn spoke]

Suno asks for confirmation regardless of what you upload, it only recognises popular tracks.

[quick sonnet]

Jus out of curiosity, which DJ?

[torn spoke]

and blocks them if it does.

[finite folio]

I don't want to reveal his name, but he is originally from Europe.

[quick sonnet]

It's OK, since you said well-known, was wondering.

[finite folio]

For the previous three songs, Suno didn't do this. for example

[quick sonnet]

Are you tallking about suno asking when uploading an audio, that you own the rights?

[finite folio]

I am a person who does not place much value on money and I value spiritual and emotional value more than money, so I did not ask him for money for this, I gave it to him to use freely.

[quick sonnet]

I was not tallking about money, just was wondering who that DJ could be, sorry.

[fast plume]

Is there a way to intentionally make Suno hallucinate lyrics when doing a cover of an instrumental?

[finite folio]

Exactly. This song called "Dream 1" and My account name was at this time a Different soul / aDs .

[quick sonnet]

I mean if it's your song, then you owning the rights, so just need to confirm it. And by confirm I just mean exactly press the confirm button.

[finite folio]

I have maded dream 1 , Deram2, dream3, Dream 4. But one song what i maded long time ago was called "Clarity"and I got inspration of Jan Michel Jarre. I maded this similar style. And during ulpoading the Clarity Suno did not aks confirmation.

[frozen dagger]

forcing hallucinations (intelligble words, like "Simmish" Sim language), in my experience, can come from generating songs without lyrics, but in the style box you describe how the vocals are sang. This can trigger hallucinated vocals. Experimentation required. Using covers and such can probably help tilt it.

[finite folio]

The Sims language, Simlish, also uses quite a few Estonian words, as it was a very exotic language for the Sims creators 😄

[fast plume]

Thanks for the tip!

[finite folio]

Do you want to try out the Exotic anguage, or did I misunderstand? Try Google Translate for example Estonian or Finnish and put the song in English, maybe it will help, because in English the pronunciation of words in these languages ​​is completely different from the actual pronunciation. And it might even sound interesting 🙂 and funny. I know because Estonian is language what is peak I speak everyday 🙂

[fast plume]

I'll try that out!!

[frozen dagger]

Very Wabi-Sabi. I dig it.

[fast plume]

I enjoy when I make a cover with Suno and it hallucinates an extra verse with random lyrics or ad libs vocals in the beginning or end to a song

[frozen dagger]

me too. it's part of my process. i like giving it some freedom and running multiple gens, sometimes trying to force errors. or sometimes you hear words differently, and it becomes part of the song after.
Opposite to trying to forge an idea, and being upset can't get it just the way you want it. Freeing and fun to explore and experiment.

[frozen dagger]

(if you have credits to burn)

[unkempt kayak]

Why is it hard to recreate the tone of the lyrics in Suno?

[fast plume]

Agreed! It's part of the fun. It gives the songs/covers some personality

[frozen dagger]

Maybe not "good" songs, by standard measure, but def not boring.

[unkempt kayak]

I've been trying to recreate this track I made using the cover feature and it won't work

Original: https://suno.com/s/s7nsYPkWWPvgjFSF

New Cover thing: https://suno.com/s/KcNIfbEFMycu4KEz

[velvet scarab]

You can try playing around with those settings to get result you want

Especially "Audio Influence" seems to affect likeness with cover the most

[frozen dagger]

my guess, would be needing to adjust sliders, want to match audio more so increase audio influence (nudge the bar). The Japanese lyrics causing Suno to pull from Japanese music it is trained on, which is warping your song to more upbeat and "happy"

[unkempt kayak]

Ok thx

[frozen dagger]

turn weirdness down to tell it not to make dramatic changes as well. (small changes to starta)

[wanton onyx]

are there more people that just get the same results over and over with same prompt with v5?
the melody structure keeps doing the same over and over

[frozen dagger]

it does happen, in V5, more than before. get locked into song tracks that all feel 90% the same. Try shuffling the wording around in your style prompt. rephrase, use synoymns. keeping prompt intact but altered can trigger changes.

[finite folio]

I change styles, for example I put together metal dance and other styles and the basic idea of ​​the song remains the same but I change it quite a bit. At the moment I am experimenting with songs I have written before, songs I have written for about 10 years.

[frozen dagger]

its as if it is stuck pulling from the same seed.

[wanton onyx]

i already tried that unfortunatly no change, yesterday everything was fine

[wanton onyx]

yes excatly this, i tried allot but i am just wasting my credits

[finite folio]

The cover leaves the music style, etc., the same but changes the style significantly.

[frozen dagger]

i speculate it has something to do with traffic on site. I also think Suno uses the workspace we work in to decide how to craft. My own conspiracy theory, but feels like clicking like on tracks influences future gens. Probably just apophenia again.)

[wanton onyx]

i thought i was going crazy but thanks

[frozen dagger]

meaning. you could try to create a new workspace and create in there, see if you notice change

[wanton onyx]

i am now trying synonym prompts. seems to do the work

thanks man

[frozen dagger]

earlier in chat i was saying i was trying to get [Maniacal Laughter] and [Laughing Maniaclly] to work and was hit or miss in past with earlier models. I got 1 track out of dozens to get it laugh using that tag. BUT, just using [Laughing] works like a charm. So, lesson here is sometimes less is more.

[finite folio]

If anyone wants to listen to a poor quality original song made with Fl studio about 15 years ago 😂 and the same song with Suno, let me know

[frozen dagger]

i tried this with a crappy 20 yr old youtube video of me playing bass guitar. I hope yours was better than my experience. XD

[fresh wave]

are any of you guys good at ending a song halfway? I have one I want to use but it just goes on about twice as long as it needs to

I'm not really experienced when it comes to, like, mixing.

[finite folio]

https://suno.com/song/5fb1a42f-fe1c-4360-9bf8-693776544e60 15 years ago one very weird mix 😂 nd woth suno regenerated https://suno.com/song/51b14046-41e3-4cfe-a287-f34a9bff582d mix 15 years later

[frozen dagger]

soudns like you need to articulate your BPM/Tempo in the style/ use [tags] in lyrics to help guide the cadence?

[frozen dagger]

Well... that is... good golly miss molly. It used the stereo panning and reverb/delay effects from the original to pretty good effect. But it went from this experimental Aphex Twin-like sound that was built from odd sounds, to some full blown Rabbit in the Moon full blown EDM,Trance,House kick. o.O

[finite folio]

I just discovered that Suno can't do a cover that has a counter-rhythm. 🙁

[proper sage]

How can i get my songs to have longer musical intros?

[frozen dagger]

can iether create an instrumental to start and d then extend with lyrics after, or try using [Instrumental Intro] or better worded taggin at the lyric prompt start

[finite folio]

I put example that
[intro Instrumental]
[Bridge instrumental] for example

[proper sage]

ahhh i've been trying [musical intro]

but i should probably specify instrumental

[finite folio]

Usually its work for me that way
[Intro Instrumental] and when music finish i put finally [Outro Instrumental]

[desert vector]

so is it now possible to put our own acapellas in suno and it changes to a different vocalist?

[frozen dagger]

if you upload a recording of you singing you can use it to create new stuff, but it doesn't replace your voice with a new "voice". Just use as influence to create new sounds.

[desert vector]

smh

check that guy using suno

[frozen dagger]

that link just opened this discord chat in a browser window

[desert vector]

its instagram

https://tenor.com/view/dumb-huh-stupid-idiot-gif-5042504

[half dragon]

probably not see the extra "dash" in the url -

[frozen dagger]

does replace voice then looking at that. ive not gotten it to keep any kind of melody, only transcribed to text. Same thing I could achieve typing the lyrics in.

[desert vector]

right yeah thats what i thought also

suno is powerful wtf

like i swear this will just COOK the music industry

they just dont know yet and see timbaland and are cooking him

[frozen dagger]

i tried months ago. and that clip is using a small example. wonder how well keeps with a longer sample

[desert vector]

yea

i just uploaded my own acapella of my own song unreleased and it said this too lol

def original too i typed the lyrics into google nothing appears

[frozen dagger]

people running into that with text too and with small common words and phrases as well. can we publish works with uploaded audio to Suno now? couldnt before. (at least if it detected vocals)

[desert vector]

interesting

[frozen dagger]

"Swift" triggers it for example

[neat nexus]

What melody did you sing it to? Try checking that too.

[desert vector]

good thought. i ended up trying again and it worked so shrug

[elder dew]

So is there a way to somehow prompt suno to NOT add the Ewwwwwooooohhhhhhhh oohhhhhhhh.. ahhhhhhaaaaaaaaahhhhhhhhhhhhhhh's at the start of your songs? It's literally happening on every single song I gen. The genre does not matter.... even happening in death metal songs lol

[frozen dagger]

if your lyrics prompt has an [Intro] tag try removing that

[elder dew]

It does not

Here is an example - I used death metal just to see if genre matters.... does not seem to, happens with whatever genre i try. https://suno.com/s/Dp4ma9F8dEqsQXyc

[frozen dagger]

doesnt look like a remix, so assume original gen with no audio influence. still looking at displayed lyrics and not your prompting, but... that's where i'd look. v5 could be part of this. seems to be something more people see now. intros and throughought

[hollow hare]

It seems to be the model itself AFAIK. I would say the choices you have at the least is either starting off in 4.5 then extend into 5.0 or remove the vocal track when rendering/mastering off the site

[elder dew]

Yeah not a remix - Original gen, and like I said it does not matter what genre I put I'm getting it close to 100% of the time.

[frozen dagger]

model is heavily biased lol

[elder dew]

just did 40 gens, lo-fi, Dark Metal, Djent, Gangsta Rap, RNB, Hip Hop, Symphonic, 40's Music, every single one had it lol. oh well. Hopefully they fix it soon

[hollow hare]

Hmm I would say the worst case option is to make an instrumental-only track then extend where you want.
Downside is that it will cost you credits but, positively, you can use various instrumental/style tags and crop the portion you loved.

Much cheaper to vocal split the starting in an external tool IMO, or using Studio (though I haven't used it as I am a mobile user)

[elder dew]

ohhh i think i figured it out -

I added in "Melisma" in the don't add "styles" Did 2 gens and nether of them had it.

[frozen dagger]

that was most def it haha. negative prompting; gotta hate 'em

wait u added melisima in negative and it actually worked?!

[elder dew]

yeah

Testing some more gens now with it in

[frozen dagger]

im shocked the negative prompt worked more than anything

[elder dew]

Yep 12 more gens and it only happened on 1

Seems to have worked maybe

[hollow hare]

The issue with negative prompts is that they are like the IT department: wonder why it is there when it works, why it is there when it fails to do what you want.

One thing that is unknown to most users is that using negative terms does bias the pattern selected for generation. Because it is looking for something specific, if your style does not have a good probability of having it, it will is more likely to incorporate the term you excluded #📚┃suno-school message

It is kind of like looking for quarters in a pile of money. If you have a pile of coins, it makes sense and has a good chance of having quarters, so it will work in removing the quarters from the pile.

If you had a pile of bills, it would be like picking up the rare quarters present (or hallucinations it being there) and showing it off

[elder dew]

In this case, I was getting Melisma on every single gen, no matter the "style" of music. It didn't stop until I added it in the negative prompt.

damn spoke too soon, now it's doing it again hahahahah oh well. Broken is broken i guess

[frozen dagger]

We are not in control. There is no spoon.

[elder dew]

no spoon indeed. Back to post production editing on all the songs hahaha. Oh well... things could def be way worse.

[ashen carbon]

hey all of a sudden I can't seem to generate with a Persona and Inspirations at the same time. am I the only one?

[fluid sentinel]

Any prompting tips for instrumentals without drums? I've been pretty specific with the genre tags but still getting beats added.

[wise kindle]

you could try adding percussion in exclude styles. I wonder if that would affect celeste and triangles lol

[rigid mortar]

Last two or three weeks it's been nothing but trouble. I don't know what they did. Now I'm just getting glitches I'm not even getting renders.

This is on every single thing regardless if I clear my cache or not or use a computer or a phone. Same thing happens every single time. A few seconds of a song followed by repeat it pieces over and over again before it just goes out completely

[wise kindle]

@hollow hare you should totes pin this: #📚┃suno-school message
seems to come up a fair bit

[elfin isle]

Yo gang. Any tips for changing lyrics but keeping the melody the same, and without changing the instrumental?

Everytime I edit a lyric I notice that the underlying instrumental is also being regenerated. Is there a way to make surgical edits? I’ve been found editing one or two words at a time keeps the instrumental closer to the original, but the words often come out unintelligible… any tips would be appreciated, as I’m not having too much luck.

Also adding to this: is there a way to “add vocal” to a track that already has an “added vocal” but select where one wants the vocal added in the timeline?

[frozen dagger]

last part is confusing, but sounds like both questions are same answer. it involves seperating the vocal and instrumentals and creating new vocals and stitching together the ones you want together with Studio. (if available) or another DAW editor.
Editor and recreate are all going to do what you are seeing and replace more than just the one word in a lyric.

within Studio (or whatever DAW) you can place the vocal stem where you want it with precision

[silk thicket]

New Jazz x Brazilian Funk Type Beat

anyone knows whats the prompt to get this type of beat? cause when I promp a brazillian funk it gives me those festival tyle lol, also for new jazz, its sounds old jazz XD

[potent cairn]

Forgive me if this asked before but I'm new here, but I'm trying to create one song that includes female vocals and male vocals to create a like the parts they sing are like from radio coming because it's a song dedicated to a movie and I'm trying to add well known phrases into the song, but since i can't get the effect of radio I'm happy to do it as is, as long are different from my main singer voice but my main problem is that while I'm using the female as intro, either goes the rest half as female singer and on other verses combined (female and after some lines male) or if it goes well on the beginning and start with female and follows the main singer, whenever the other male vocals are in order to appear it goes with female vocals again and it continues with a duet female and after some lines male.
Is this a bug or something that i can fix because i have tried at least 30 adjustments and generatings.

[quiet solar]

Don't worry, it only gets asked about 4-5 times a day.

It's recognised to be one of the more difficult types of things to do with Suno, and there's no guaranteed way to achieve what you're looking for, but if you search for "duet" or "male and female" you'll soon find the advice people have given to try to tilt the odds in your favour.

Expect that it could take a lot more than 30 generations to get right, though - especially if you're trying to do it all in Suno.

[past jewel]

Do one of you know how to stop snares entering the songs 😛 It's so anoying to make drumloops. but i dont want any snares in. exclusions are set. but it keeps ignoring the exclusions.

[unkempt sinew]

That's close to impossible to have Suno do 🙂 I would rather just let them be part of the track but stem out everything and remove snares there

[raw tiger]

Is there any sort of free definitive (or the closest thing to it) list somewhere of both Custom-mode bracketed lyric prompts and styles prompts? I am having a tough time getting the sound I want, especially when it comes to building to a crescendo in alternative/indie songs and certain pedal/guitar/and mic effects (but only on certain parts of the song)

[raw tiger]

You would think that since every art app under the sun has a negative prompt box (where you write things like “extra fingers, unrealistic noses”) that Suno could too, no?

[finite folio]

Hello 🙂

I made a song that I made in the original and is 6:22 long, When I bought the pro it was marked there that it is possible to make songs up to 8 minutes long. But quite a few songs are 1:20 to 6:22 minutes long as a cover, the songs in the original are longer than 4 minutes. Why does Suno cover so many shorter songs from the original, when when buying the pro it was said that songs up to 8 minutes long are possible.🤔

[wise kindle]

it can generate up to 8mins - I've had a few 7:50+ songs - but it's rare

[past jewel]

you could try "extended full-length 8 minutes" i like to use "radio edit 3 minutes"

[wise kindle]

and honestly, I deleted them. it pads them out with boring stuff and repeating patterns and wasn't at all interesting.

[finite folio]

When uploaded, the original is, for example, 6:22 in Suno, but Suno does a 3-minute cover, for example.

[past jewel]

on the remaster?

[wise kindle]

a cover is just a song inspired by it. you're not guaranteed to get a similar length

[finite folio]

I put there make cover. and i write 1:1 cover of this music and styles.

[wise kindle]

lol you crank up the audio influence if you want something closer to the original. you don't put 1:1 cover in the prompt

[finite folio]

I also mark the length, the same length as the song or instrumental track, in that box.

[wise kindle]

it also doesn't work like that. with AI you basically get what you get. You don't have that much control, no matter how many text boxes and sliders there are available

the only way to almost guarantee a longer song is to use a well-known slow style and try andante or adagio tempo (it does recognize tempos pretty well). but of course this means you're just slowing the song down, not really making the actual material longer

[finite folio]

But for the songs that worked, the melody and spirit have remained the same and Suno hasn't added much of his own inspiration. He's just overwritten them and the quality has improved.

[wise kindle]

Even at ~20% audio influence I find it'll still be pretty close to the original, at least structure, rhythm and melody-wise. Maybe try dropping it a little more?

weirdness can influence it a lot too. the more weirdness the more creative it will try and get

[raw tiger]

is that a no or a stupid question?

[finite folio]

What is the "Advenced Options" in Suno? Is this to mark what Suno would leave out of the music or songs? I still don't understand this choice.🙄

[wise kindle]

It allows you to exclude styles (like pop, rock, etc if you don't want those). You can set the gender of the singer, then there's sliders for adjusting the weirdness (can add some pretty cool complexities to your song and fun sounds, but don't set it too high..), the audio influence which is how close to the original it will be if you cover, and the style influence which affects how much the style prompt influences it

[last bay]

negative inspo would be nice to enhance options

[near flume]

Try [Intro - Instrumental]

[finite folio]

I also haven't figured out the sliders, if I choose female there and want a lower voice for a female singer, which way should I move them? Logically, should I move them to the right for example a higher soprano and to the left for a female voice with an amaladm, for example a kotraal, and for women the same thing, that a higher tenor to the right maximum and a bass voice to the left maximum, for example, or what the hell is with the slider (I haven't dared to try it myself yet).

[finite folio]

I have used [Intro Instrumental] and in the advanced option I have written vocal, voice, and I have also sometimes put the Style in the description -Voice and violence prohibited for example, sometimes it works very well, but be patient just in case 🙂

[south shoal]

I'm having similar issues. Between one in ten and one in 20 attempts provides something that can potentially be used. When a Cover starts that I like, I'm hoping it doesn't play up. Most of the time it does and I have to spin the roulette wheel again. On the positive side, Suno has delivered a few outstanding songs that I love. On the negative side, it burns credits and can be frustrating. I suppose we have to be patient in the hope that over time these glitches become fewer and farther between. Changing the slider settings doesn't seem to make much difference as I thought that may be the root cause of the problem.

[finite folio]

How does Suno calculate the tempo of a song? I uploaded a weird mix of mine that I made years ago to cover it, and the styles tab says, "The song has a fast tempo, a 4/4 time signature, and a minor key signature." Actually, the song isn't 4/4, it's slower. it's medium but not fast. its more like moderato
90–110 BPM

[frozen dagger]

dunno if you saw Poltergasm mention the exclude after you asked. but there is an exclude in the advanced options. It is still hit or miss (like a lot of a.i. generation), but it is there.
#📚┃suno-school message

[finite folio]

Currently, Suno offered one style assignment as the last line in the Style box of my uploaded song: "There are no vocals in the song.". Suno has never added this before, this is the first time I've seen it.

[finite folio]

Okay, I don't understand Suno's tempo calculation, 70-140 BPM 4/4, it's the same tempo for everyone, there's no difference in tempo or speed.🙄 70 BPM 4/4 should be a much slower tempo than 120 or 140 BPM 4/4. Or maybe I'm old and logic is failing me.🤣

[quick sonnet]

Possibly that negative prompt can help, but since I am not a Pro user can't confirm.

[finite folio]

I tested with metronom this my greated original music track there have 82 BPM and cover of this original what i set 80 Bpm have 140BPM 4/4 not 80BPM i used metronom for that.

[last bay]

For deeper female need to prompt and when you get a good one make a persona

[thorny flicker]

Does making a persona keep the same voice? Cause I tried making a few persona and there's very mild similarities, except for the voice

[last bay]

Can also try genre that is more likely to be deep for the persona. But then night need from stem or a small part

[finite folio]

A metronome is a good thing. Many thanks to Johann Nepomuk Mälzel, who invented the physical metronome, which is still widely used today. 🙂

[last bay]

Not exactly same is also more elements from what the persona is made off

As said Can help to edit a song to a small part and make persona from that instead of whole somg

[finite folio]

Hallöchen 🙂

[last bay]

Hi 🙂 hm I think I hid all my aggressive females with deep voice unpublished 😄 maybe time for another take

[frozen dagger]

A dark ambient soundscape with a slow tempo and a minor key, The piece features a sustained, low-frequency drone that creates a sense of unease, A distorted, almost guttural vocal sample is layered over the drone, repeating a phrase that is difficult to discern, The overall production is sparse, emphasizing the unsettling atmosphere, There are no discernible chord progressions, only sustained tones and textures, The vocal quality is heavily processed, suggesting a robotic or monstrous origin, The song structure is minimalistic, consisting of a continuous loop of the drone and vocal sample, The melody is non-existent, replaced by textural sound design, The tempo is extremely slow, contributing to the oppressive mood, Production elements include heavy distortion, reverb, and a low-pass filter on the vocal sample, creating a muffled and distant effect, The mixing emphasizes the low frequencies, giving the piece a heavy and resonant quality

I uploaded audio that gave me this prompt. Did a couple gens and it reminded me of this. There was some small drum near the end of on gen, but with no lyrics and just this prompt, and nothing else, did pretty well with creating a percussion-less track. (with no negative prompting.) Would modify it to personal needs, but can find what is working within the prompt to reproduce the "drumless" sound.

[frozen dagger]

if you have a sample (not copyrighted) try uploading it see what Suno gives it as a prompt to look for clues to prompt.

[last bay]

Hm could try it with an old song I never could quite capture the vibe in v4 upwards

[safe cliff]

So Suno can't "Get full Song" anymore?

It s an option for v.3 generated tracks, but v3 isn't available snymore

[vocal owl]

Hey everyone 👋
Quick question — is there a list or overview of all the available prompts or commands that can be used in Suno?
I’d love to see what’s possible or get some inspiration for creating songs. 🎶

Thanks! 🙏

[raw oasis]

Mad Scientist rules in effect.

[calm karma]

not a prompt thing but no real section for this:
so i have been running some tests with a song I created. It says its 91bpm, but I pull it into my daw do a beat detecttive and it says 120bpm. Now to make it worse if you line up the track then throw a click on it some parts its right on and some its way off. Cant figure out whats going on?

[raw oasis]

Dynamic tempo can be good sometimes though right? unless it is super wonky.

Had one yesterday I put in Tempo Track in Studio One and it varied through parts 71-90bpm

[calm karma]

totally agree. makes it tough bringing it into a daw and writing midi

[raw oasis]

Yep almost have to midi paint piano roll

[calm karma]

i cant figure out best way to get after this. Like I want to redo the drums cuz there are a lot of artifact issues and volume issues, but if I cant get the midi to line up. No idea what to do

[raw oasis]

I do sometimes have success using NeuralNote - take a stem and process it out to MIDI. user results may vary. Throw EZ Drummer on it. adjust.

Ideally Suno would give accurate BPMs for our songs.

[calm karma]

yeah ive racked my brains with that. In truth Sunos stem spliting is noisy and misses half the notes when transfer it to midi. Ive tried many different midi converts. Not NeuralNote though. D

do you have alink for that?

[raw oasis]

Studio supposed to let us redo just drum tracks.

[raw oasis]

v 1.1.0 is latest release. https://github.com/DamRsn/NeuralNote/releases/tag/v1.1.0

I am gonna upgrade I think. treat myself for a month.

to Studio.

[calm karma]

yeah i have studio. And you can re-roll any track. Its actually very nice. But I havent figured out how to make it play the exact same thing only clearer

let me know if you have any luck with it. Would be nice to have a group of people that help each other out with this stuff

[raw oasis]

okay sounds good.

[calm karma]

hey I installed the standalone for windows, but I can find it to launch it. Any chaance you know the install location?

got it