📚┃suno-school

[unique stone]

Your screenshot shows you were previewing a replacement. You need to press the "Commit" button.

[raw orchid]

The last one I did, it didn't put it in my workspace, but it was in my library

I used the "got to song" that appears towards the center left of the top bar after "Save as new song" is used

[polar spade]

Yes. Bought extra credits attach to the account, not the subscription. So they are good always, so long as you dont cancel out of Basic.

[pliant idol]

extends are broken

[polar spade]

I always have to go back to desktop to check, but , even though we're not literally previewing anything, we still need to click ESC to drop out of the popup first, right? Else we can not click anything at that point,until we do. "Stop Preview" is showing too.

Besides, its not like the previews will disappear, unless we hit Reset All.

[paper dawn]

I think is a limit for IOS for total amount of songs/editsin one workspace or maybe I don’t know how to change on iOS but I edit a lot and once I got to a certain amount my early songs started disappearing I can still access them on the computer but I can’t on iOS there’s nothing I can do to see a different workspace or an old song that I’ve made.

Yeah I know I shouldn’t have that much but I edit everything I edit a lot sometimes I edit stuff that I shouldn’t but that’s on me , but still I can’t access it on iOS . I would love for a download everything button and you guys probably save some money on hosting if I got all my edits off your servers lol

[sleek stump]

Just a suggestion... if you are doing a ton of editing, and then complete your tracks, and you really want to save your sections and snippets, dload them right away and delete them from the platform. Mobile apps have a ton of limitations, and with that many clips and tracks, it's very burdensome to navigate the library. Personally, I clean out my workspaces and library at least once every week to two weeks. It got really frustrating for me when I got to 15 pages of stuff in the way and 15+ pages of trash.

[paper dawn]

Yeah I know that’s what I’m saying give me a way to get it all off cause I do want it like even before I was on the AI any song I made I stripped down the music and the edits for either samples or loops but yeah it escalated very quickly

[sleek stump]

YOu can multi-select now and use one action to trash all at once, and in trash, delete them pages at a time

Well,in the browser page you can, not sure about teh mobile app, I only use the mobile app as a car playlist

[paper dawn]

I haven’t seen that I’ve been waiting for a multi select how does that work I’m looking at my computer now. You mean through the filter filters and just do it that way?

Not me I’ll make a loop on splice I’ll take it to freaking in track or something else and then it’ll eventually come here like I love having everything at the tip of my fingers and it’s easy to make something way nobody really knows you’re doing it

[sleek stump]

Looking at your page of tracks, hold CTRL and click to select all teh tracks you want to action, then hit the more menu and select trash (or delete if in the trash folder).

[paper dawn]

Oh yeah and then just shift and hold down I didn’t I didn’t think you could do that For anything other than trash but I see I can at least move it to another workspace so I can access older stuff now I appreciate that trying to figure out if I can download it all right right now cause I will

Yeah I gotta download it all first cause like I really do edit stuff that doesn’t need to be edited and then I forget about it I’ve made songs that I forgot about I’ve I’ve done a lot of stuff that I’ve squirreled out and went on some other freaking mission and I really would have to sit down and go through it

Most people get 2000 songs a month with their monthly subscription I get about 10 or 15 songs through edits and I spend every credit

[sleek stump]

Ya thats a handy thing they added.. you cant multi-download anything though, they use streaming dload which doesnt allow that.

Thats why anythign I want to keep I get it quickly.

[paper dawn]

Oh yeah you’re a genius I can click on the filter for full songs and then do remixes covers and edits and put them all on us you just made me look like I’m actually somebody who’s organized

Tutu

Tyty

[polar spade]

.

@fading tulip I suspect, due to the weekend and holiday, that it could be Monday before communications are reviewed, from all platforms.

.

@oblique fable I see you have at least 2500 credits when signing up as a Pro member. Additionally it appears you may have bought extra credits July 5th and 6th? That seems to be the additional 330 credits as of the time of screenshot? But are the 2830 ALL extra credits?

@oblique fable Ahh. You HAVE credits but your account is locked/in limbo. Do I read correctly that your payment will not process from YOUR account until July 7th, tomorrow, (or today, Vietnam time)? But even if it removed funds from your account, international bank transactions can take up to 4 banking days for new transactions for an account to post the payment

[polar spade]

@gentle lotus #archived-ticket-help message

@arctic beacon We get subscription credits that refresh/top up at each renewal monthly. They dont roll-over. We can purchase extra credits (starting at $4 for 500 credits as example) which never expire.

@titan.jim @mighty bane we can not buy a license to use outputs Suno created for us while we are in a free subscription. Those belong and remain with Suno. You can re-run the whole lyrics/prompts and see what new output you get while a Pro member if you like

[steel crescent]

Hello, I was wondering if anyone else had the same issue as me.
I extended my song to get a better outro, but when I generate the whole song and listen to that specific part, the vocals turn into chipmunks and it's completely unintelligible.
It’s the first time this has happened to me on Suno, and even if I regenerate the whole song with the same extension, the bug keeps happening 🤔

[sleek stump]

So, when you "Get Whole Song", the resulting new track has a different extended section from your generated extension?

[steel crescent]

Yeah, all the sung part is gone into... 🤷‍♂️ It's completely unintelligible

[sleek stump]

But the actual extend is ok by itself?

[steel crescent]

Yeah, that’s why I was wondering, it’s the first time it’s happened to me 🤔

[sleek stump]

Sounds like an issue with the web service backend... it should just put your original and teh extend together, but it sounds like its doing something unintended with the extend before it stitches them together.

[steel crescent]

Yeah, there’s definitely a problem with “Get Whole Song” because I tried it with another extension, and while I do get my outro, it actually cuts off about a minute from the extension 🤔

[sleek stump]

HAve you tried a second time?

[sleek stump]

Was that within a short time frame?

[steel crescent]

Wdym

[sleek stump]

Both of teh error states were close together or were they hours or days apart?

Trying to determine if escalating through a Mod is prudent

[steel crescent]

Oh the first time i tried was like 20min+~ and i just tried again right now

[sleek stump]

I'll do a test.. are you useing teh new editor or the legacy editor?

[steel crescent]

None

lol

[sleek stump]

SO yo uare just using the OG extend function? ((from teh options menu)

[steel crescent]

This is the extend https://suno.com/s/B0An0LbeOo9fZf9h
And this is the whole song https://suno.com/s/E95j2ydWU9b1jJvn Outro should be around 3:20~

[unique stone]

I'd be interested in looking at this if you can post or DM me links to the final part of the song and the "full song" along with the specific time that you hear the problem. It's unlikely I'll have a solution but I'll see if it's similar to other issues I've been looking at recently at least.

[steel crescent]

Yes i do

Wait am i drunk 🤔

Im pretty sure the first time i listened to it it was not that

Welp 🤷‍♂️ Gues it's fixed

[sleek stump]

Thats sounds the same to me

I dont hear any difference

[steel crescent]

Yeah me neither, but i still tried to gen the whole song 3 times, so i don't know if im crazy or 🤷‍♂️

[sleek stump]

I put both in my DAW and they are the same from teh extend point

[steel crescent]

Sorry for bothering i guess
Maybe i'm tired or maybe someone is playing with my mind 😮‍💨

Thank you guys anyways 👌

[unique stone]

Is this for a different song / is this one still a problem?

[sleek stump]

Just did a test extend and didnt see any issue... so I dont think teh platform is having issues with this currently

[steel crescent]

Apparently i was the problem, sorry for bothering 😔 And it was for this song

[sleek stump]

Have a beer and enjoy the day 😄

[polar spade]

I mean, I heard the same (reverse-turntable) scratching in your sample and in the full song at 3:15. And then the wah-wah chipmunks ( vocoder) sound at 4:20. Both just sounded like typical funk groove moves to me, like Parliament would do.

.

.

[polar spade]

They may be renaming features. Yay!! I want in on this. https://discord.com/channels/1069381916492562582/1290756014714519683

[boreal briar]

hello ive get an idea to use suno to make a radio drama. ive written the script and and now im working on the structure and prompting but as there is no resource that show what the model use and understands ive used the same techniques as i world with a song and ive close to what i want but there a lot of inconsistently and i have no idea how to get sound effect to work. if you have any advice i would appreciate it.

[sleek stump]

Being that Suno is not trained on sound effects, its going to be difficult to get sound effects and even harder to get any consistency. Its also not easy to just get any real length of spoken word or dramatized spoken text. I have had some limited success with compltely spoken tracks, but it took a lot of gens and adjusting of prompts.

[pliant idol]

Get Whole Song feature is currently broken with errors on all OS.

[sleek stump]

Additionally, Suno is a generative AI, and it has a very high amount of random generation that is native in the model.

[sleek stump]

I just did a test of one 10-15 minutes ago, worked OK for me

[pliant idol]

@sleek stump - try it now.

[sleek stump]

Running another

[unique stone]

Seen a few of these over the past few days and have been able to reproduce it. Until it's fixed, a workaround is to make another extension off the latest part.

[pliant idol]

any Get Whole Song fails with an error...does not matter the extension.

[sleek stump]

Worked agaion for me 5 seconds ago

[paper dawn]

Yeah you actually have to either heal your edit or you have to do that and align your lyrics

Sometimes it does just mess up and it’ll take out a whole verse but like if you’re editing where you come off if you edit into a verse it’s still thinking that that verse is all the way being played out you get what I mean so if you have any kind of prompts or tags it’s gonna it’s gonna do something differentthat is unexpected

[sleek stump]

He wasnt using the editor.

[paper dawn]

Yeah but it’s still technically editing because I do a lot of extensions when you extend if you extend in a verse that is already made on that song and say you go into a tag or into it like the last word even

[sleek stump]

YOu cant do anything with the OG extend window

[paper dawn]

What do you mean just clicking extend? And not using the new editor?

[sleek stump]

I use teh OG window for extends because its alweays been the most reliable for me.

Its a version back from even the Legacy editor

[paper dawn]

I do multiple edits like every song has at least like 20 and one thing I’ve learned is when you edit you have to to make sure that when you replace your lyrics in the next window that other lyrics aren’t still in there from the last verse because it’ll put them together then it’ll sound like they’re not even speaking a languagebut it’s all about the way you’re editing from what I’ve understood like I’ve been able to stay away from that for the most part but it does happen but it’s because I got lazy

And then you have to line them too sometimes that don’t even help

[sleek stump]

Yes, this is all good for the new editor, but the OG tool doesnt have that for extends

[steel crescent]

Yeah, everything's alright 👌 I was probably just too tired, too drunk, or whatever but it's all good now. And I use the OG extend because I’m old and don’t like when things change. 👴

[paper dawn]

lol well then I didn’t even notice I was using the old one I just click extend on the

Song

iOS using new or legacy?

I guess it really doesn’t matter I got it down to a science that if I wanna persona or if I wanna replace a part of a song I use my computer everything else I can pretty much do on the iPhone

[sleek stump]

Mobile apps are a version of new that are still not as feature rich as teh web UI.

[paper dawn]

Yeah and I figured that because that’s how every app is it’s not as advanced as the website is usually

[sleek stump]

There are a lot of limitations just due to the processing and GPU power limits of mobile devices

All the stuff you do in teh mobile app happens primarily in your phone, not on a server in a data center like teh web app

Ajnd teh mobile apps have features that web do not, and those require resources on the phone also

[paper dawn]

Wait it’s our devices using GPU or is it using a virtual machine of Suno?

Oh OK you already answered that

That’s wild but that makes sense why my computer takes forever

[boreal briar]

i know A.I works best when you keep things simple so the drama was written using the same format as a song (keeping dialogue short)the results haven't been that bad, if i can brute force sounds im might be able to edit everything together. thank you

[sleek stump]

Genaral internet has lots of issues with periodic latency, packet loss, failed hops, ect, all that can affect how fast or slow the web UI is.

[sleek stump]

Good luck with your project, hope you find success!

[paper dawn]

So why don’t they just save their servers and make everybody save their stuff on their own machine that’ll even stop some of click degenerates (click to generate) but that would save you guys some I mean it’s not gonna save your GPU or CPU or memory but you could probably trade out some of those SSD for more 4U servers lol

[sleek stump]

Hehe, those are questions us users just cant answer 😛

[paper dawn]

Oh I was almost talking to you like you built it yourself you’re pretty smart so lol my bad

[sleek stump]

Been using the platform for a while, and like a lot of long time users, we have come to understand how a lot of the platform works

[paper dawn]

Yeah I get it but that’s pretty cool you take it serious and you take pride in it and you know what you’re talking about

[sleek stump]

4.5 has been the biggest learning curve to date

[paper dawn]

Oh snap means you’re getting used to it that means 5.0 will be out very soon

[sleek stump]

Theres a few of us that hang out here and try to help folks, all long time users.

[sleek stump]

lol.. I feel like I have only scratched the surface 😛

[paper dawn]

Well I’m appreciative of you you already helped me today and I went to school for a computer so I appreciate you 100%

[heady pebble]

Has anyone figured out how to dial in a specific tempo at all or is it a "it'll pick a random tempo from the style description"
Trying to make some faster songs with styles that normally aren't fast and the normal tempo prompts I try don't seem to affect it.

[sleek stump]

Traditional tempo terms Suno understands. it does not seem to understand BPM.

You may however have a hard time trying to gen high tempo songs for genrte that dont normally have them. You may want to increase speed in teh editor.

[heady pebble]

Yeah I ended up changing the BPM in the editor. At least the song I wanted to manipulate was an instrumental test so the vocals wouldnt go all squirrel.

[sleek stump]

Theres also a check box to keep pitch, so even if it did have vocals, they would be faster but not chipmunk.

[quick agate]

hey, so uploading songs - while cool that it looks for existing material, im looking to remix a song that IS in public domain, its well over 90 years old so not sure if I can appeal the upload or something, I mainly want to extract and work with the 'tune/jingle'

[spare musk]

i need help, can i change a female voice to a male one after my song is created?

[polar spade]

Yes. It may be as simple as doing a cover, putting "male vocal" as your first style prompt, sticking [male vocal] in the beginning of the lyrics and rolling the dice. It can work.

Of course just clicking create 10 times may also randomly get a male vocal thrown in as well. Its not so unusual to not be surprised if it did.

....and just post How-To questions in here is fine. No need to double post also in the need-help channel. Thats mainly for billing and account questions.

[spare musk]

sorry my bad

removed 🙂

thk you for the answer

trying to do a anime style song but its hard to alternate voice

[polar spade]

Well, the weekends, that channel has few mod or users to visit as well with a 2 hour re-post delay. Its a mostly Mon-Fri thing.

[polar spade]

I use extensions with personas more often. There are other tricks but all depends on your timing and setup. If just singing verses, extending with personas can be a helpful option to change voice and style too if preferred.

[spare musk]

good idea

weird i regenerate a part and the lyrics so weird i cant understand what she is saying ahah

[polar spade]

Just as proof I took about 30 minutes yesterday and ran through a song changing through 6 , maybe 5, I forgot, voices and styles by doing extensions and adding personas at each verse while still maintaining the motif of the song.

[polar spade]

If you add personas, suno likes to throw in melodic interludes between the verses, so then we're looking at splicing things out here and there.

[spare musk]

yeah i got a nice personas with duo voice , work pretty well

im so fan of switching voices

as a fan of metalcore 😛 scream

[polar spade]

Try it out. See how it works for you.

[spare musk]

its working, i will show you after my song will be complete if you want

need some feedback 😛 its a "one piece song"

[polar spade]

Im terrible at feed back, but you can post here and ask anyone to give a listen. Good luck, mate.

[sleek stump]

Didnt see any response to this so I'll add it.
Suno does not allow uploads that are not completely yours, you must own and have all rights to the content you are uploading. That is why the popup box is there when you upload.
In essence, you can not upload a recording of anything you dont own and do any cover or transformative work on.

[deft coral]

Please advise how to reset the SUNO “memory.” SUNO has “remembered” a couple of generation options and now only outputs variations of them.

[limpid grove]

i've got a 8 sec audio sample, which takes a very long time to upload... any known issues with audio upload function???

it's like it times out while uploading....

[sleek stump]

It could be due to the content filter detecting copyrighted audio, there are a number of reasons uploads can stall, unfortunately theres not any messaging saying why or a clean exit most of the time. Try a different 8 second audio of something recorded locally ans see if that fails or goes through. If it uploads successfully, its the content filter for the original you are trying.

[sleek stump]

You can try the following:
Log out of Suno and the service you are logging in through ((Google, MS, Apple ect)
clear your browser cache AND your cookies
Log back into your accounts..

This has worked for many users @raven orchid

[polar spade]

You might start with the Clear All button. Some don't realize there's one at the bottom of the Create/Remix panel. Its also logical to waste a gen on some odd song having no relationship to what you were doing, then go back and start again.

Superstitiously, I've just a habit of often giving it 30 seconds after clearing, if i think its repeating, before starting again.

[polar spade]

I've just stopped responding to those questions about public domain and asking "permission" to put the risk back onto Suno. It only puts users at risk as they've already agreed to hold Suno harmless.

[raven turtle]

WAV downloads are timing out... hmmm

[polar spade]

Seems that way, since at lease Friday (?) Some users having luck to copy the file and try downloading THAT new one.

[raven turtle]

Thanks. MP3 or stems seem OK for now, but WAV isn't budging! I'll copy and give that a go. THanks!

[polar spade]

Are stems d/l as WAV files?

[near flume]

run a generation or two of something completely outside of what you normally do. Mix 3 or 4 unrelated genres.

[raven turtle]

Yes. Oddly enough

[polar spade]

Im trying to see if its file size, length of title, any consistent thing that might be the issues. OK. Thanks for the info.

[sleek stump]

I'm not seeing any issues with WAV dloads, have done several and one just seconds ago with no issues.

[raven turtle]

Just created a cover of the sticking WAV and got a download from that! Guess Suno doesn’t like Electro-Industrial too much! 😂

[sleek stump]

I believe files are getting corrupted in the back end, and in my experience, and after spending months and multiple holidays in here, these issues are more prevalent when the service is under heavier load. A lot of people (including devs) say thats not the case, but its too coincidental over many months.

[polar spade]

That certainly makes sense. I've yet to be able to repeat the problem, all my downloads are fine, large files, small, long titles, whatever. So it can very well be non-specific.

Someone in bugs mentioned they did an extension , Get Whole Song, then could download new WAV then. So same things , basically. Kinda different song, but, the idea is there.

[slim kelp]

Hey It just happened, I just generated a song that can't be downloaded in wav format. no any remix and something, just input my Lyrics and click download wav. other song downlaod work fine.

[sleek stump]

This might be more evidence for my hypothesis...

[polar spade]

It shows that the "preparing your download...." had completed, but still got an error once you click Download File?

I'll assume many may simply be clicking "Download File" again and thus getting the same error. Anyone here try closing everything and starting at the menu and download the same file, to see if its something to do with the processing for download?

Making a copy of the file could be a red herring if most are just clicking Download File over and over again without starting over. Just going over any possible scenario. The things we do on holiday.

[slim kelp]

I think it's a UI bug

[polar spade]

Did you back out to the beginning and try downloading the same file from the start?

[slim kelp]

its load infinity and show error

Yes, it will show a timeout.

and show this panel

[polar spade]

It goes so far as to give the Download File button, but then times out and if its finished processing, what it is timing out on? Yea. totally strange then. UI issues, for sure..

[sleek stump]

I've never liked streaming dloads...

[viscid badger]

Hello everyone!

I have been trying for some time to generate a duet of a woman and a man.

The meta tags currently say:
[[Male vocals, Verse 1]]
[[Female vocals, Verse 2]]
etc.

Some of these verses are sung correctly, others not at all - suno has real problems with this.

Does anyone have any tips?

[sleek stump]

Are you detailing the duet in the Styles window? Also, what model version are you using?

[slim kelp]

it will show the "download button" after timed out, when click this button will show "Download failed, please try again"

[viscid badger]

Model v4.5

The style window says:

A duet featuring piano and subtle strings sets a melancholy, profound tone. Soft vocals from both artists intertwine on the verses, building emotional intensity in the choruses with layered harmonies. Sparse percussion adds weight, while a cello-driven bridge deepens the sadness before a restrained, delicate outro.

[sleek stump]

Theres part of the problem, duet is not detailed in teh styles prompt

[polar spade]

Thanks for the extra details. Its happening to many , so I was hoping for a pattern, but likely , as Ginger says, something backend we have no control over.

[sleek stump]

Here is an example of the duet being called out in styles...
A modern pop duet featuring both male and female vocalists trading lead on alternating verses, supported by warm piano chords and pulsing synth bass. A lush, full-band chorus unites both voices in harmony, with understated acoustic guitar and soft electronic textures for dynamic contrast.

[viscid badger]

The generated song has a male and a female voice, just sometimes in the wrong place. So Suno understands what I want, but confuses male and female in some verses

[viscid badger]

I'll try that, thanks!

[sleek stump]

Then you reinforce where you want each to sing in the lyrics with tags [Verse 1 - Male] [Chorus - Female]
These are some examples

[viscid badger]

Yea I did this

[viscid badger]

Same problem. Some verses that say [Male vocals] have a male voice, others don't. Likewise with the female voice

[sleek stump]

Its likely a matter of what your tags look like...

[viscid badger]

Pretty basic:

[[Male vocals, verse 1]]
...

[[Female vocals, verse 2]]
...

[[Both Voices, chorus]]
...

[sleek stump]

You could also be up against a heavy load issue on teh service and Suno will just refuse to cooperate at all. I have had this happen during holidays.

Ya those taghs are not what Suno wants'

Try these instead:

[Verse 1 - Male Vocal]
...
[Verse 2 -- Female Vocal]
...
[Verse 3 - Duet]
Dont use double brackets either.

[viscid badger]

Suno just doesn't get it

Well I guess it's not officially supported

Both the female and the male voices are so beautiful, and they are indeed taking turns, but in the wrong order😂

It's not male, female, male, female

But female, male, female, female, male...

[sleek stump]

Many people get good duets, I have also. Like I noted above though, Suno can just be stubborn for an entire day, or just getting unlucky with the RNG aspect. Also you could try turning up the Styles influece slider high and the weirdness down low.

Suno can make some amazing tracks, you just need to figure out how to speak to it to get close to what you envision.

[viscid badger]

Yea I got some incredibly beautiful songs already, But they weren't duets. Anyway, I will experiment a bit more, thanks for your time and help!

[viscid badger]

https://suno.com/s/3trTZjSAc7x4aFFH

That's the best I currently got, here suno only mixed the genders up once, the rest is correct!

However, I don't find this version all that nice. I would have liked more emotion and more beautiful voices😄

Well and the duet part isn't correct

It's a good start but I'm not satisfied yet

[solemn lintel]

Any suggestions on how to get a cover to actually be a direct cover, but just change the voice? It sort of used to work that way in older versions, but now covers completely change the sound of the song, and adding a persona in only works like 10% of the time.

[sleek stump]

Use the sliders, turn influence of audio high, weirdness low and style influence high

[solemn lintel]

Ah, OK.

That's better, though it still doesn't help the persona to actually apply. I guess that's still a work in progress on Suno's end. I was so excited when they added the feature.

[sleek stump]

Persona are not what a lot of people expect...

[solemn lintel]

Apparently not.

[sleek stump]

Most users think they are vocal presets, they6 are far from that 😛

[solemn lintel]

What's weird is the song I'm trying to "fix" has exactly the right vocalist starting halfway through. I wish there was some way to tell Suno "Hear this voice you use in the second half of the song? Use it in the first half, too."

[polar spade]

Then at least save one of those and remaster in 4.0 to , we would hope, equate all female vocals to be the same as the first one.

[solemn lintel]

WEIRD. That actually worked!!

Mind blown. Thanks!

[polar spade]

Yep. Glad it worked. (Not always works 😉)

[keen iris]

I've created a cover with my persona and this track just wasn't added to my persona, is it just have one track it use as a reference?

[sleek stump]

Persona are made from a single track/clip, you use the persona to influence new songs or covers. The new songs do not get added to the persona, if I understand what you're asking...

[keen iris]

Yes this is what I'm asking, thanx

So why new tracks made by persona can't influence the persona?
I mean tracks user chose to add to the persona's repertoire.
It'll be great feuture tbh

[sleek stump]

Persona are influencers, they are used to maintain a theme, you would not want a persona being changed by something you used it on.

[keen iris]

Well actually I do want it 🙃
Or atlist I want to create persona from cover made by persona.
I've created a cover from one of my suno tracks with persona and I can't create persona from that track.
I need it so bad, because voice there is what I trying to get for months + waste hundreds of credits on this

[polar spade]

You can do that by sliding the Audio slider up a bit more. Then when it sounds like a good blend, keep that and even make a new persona from THAT output of you wish.

Or you can make a persona from song 2 first. Go back to the original song of your first persona. Cover it and add song 2 persona, so you have now influenced the persona song by the "second" song. 😄

That is just what user have typically called "building" or "layering" a persona.

[midnight steeple]

And Persona, when I use a persona, and then try to make a slow country song or a fast country song... if I use the same persona, it uses the same melody, intro, etc... I would just like it to use the same voice and vocal style.. not the same music.

What is the purpose of audio uploads... I would be nice if it had a way of taking our voice and making it sing... Or... if I record me humming a melody I make up, it takes that and turns it into actual music... but as of now, I can't find a use for it.

[sleek stump]

The sliders will help negate those effects, you can turn up weirdness to get more complex sounds, and you can adjust the sliders for audio and style influence...

Sliders are relatively new so a lot of folks still arent used to them even being available

As far as uploads, what you are wanting is something a lot of us want, Vocal Presets... Suno doesnt have them... yet.

[primal juniper]

Can't download wav files anymore. Wav file download timed out. Pro user

[polar spade]

Likely a server issue. I believe Devs are aware. Until fixed, try to copy output and download the copy.

[spark bobcat]

Anyone having issues tonight with saving of Personas?

[meager frost]

Has anyone managed to get the sliders (like weirdness etc.) showing on mobile browsers? Either on Android or iOS?

[onyx finch]

You checking "view desktop site?"

[polar spade]

@fading tulip Community response: As mentioned before, , the general hours for Suno runs around the typical Monday through Friday schedule it seems, with a skeleton crew on weekends, Id consider the prospect with there being basically a 4-day USA Federal holiday this weekend, this would factor in as well. Your thoughts are as good as mine from this perspective.

[polar spade]

As @onyx finch mentions, clicking on "desktop site" in your phone (under the upper right menu, often) could help change the format to show the sliders.

[meager frost]

Thanks guys! Just wanted to let others know in case they’re facing the same issue, on iOS, Safari doesn’t seem to give the option to request the desktop site (at least for me), and even though Chrome does let you request the desktop site, it still doesn’t show the sliders. But with Opera, after requesting the desktop version, the sliders show up just fine✔️

[polar spade]

Yep. I use opera. Opera and Dolphin have desktop modes which will do this. Ill assume some iOS dont do this as well, then. Chrome on Samsung has the setting in its menu. You're right, some Samsungs dont offer the desktop mode in their menu though, probably same as IOS. Sorry for that suggestion if it didn't apply to yours.

[hardy thunder]

I also tested the desktop versions of Safari and Chrome on my iphone and the sliders don’t show. (At least on iphone Safari desktop can be requested by clicking the AA on address bar). Luckily I use a PC for Suno.

[polar spade]

So with either of two accounts, when I click an output in Library, it tries for 1 second then goes back to Play button. Thats odd. Seems like a phone issue. I'll restart phone maybe. Any others having issues on just phone? My desktop is no problem.

[graceful condor]

why bother having/using an [Intro] if its just going to keep skipping it? 6 times in a row. It just doesn't like me.

[graceful condor]

Skips the intro. Just tested it happened When I have other tags before the intro like this [Title: Where Mercy Breathes] [Genre: R&B + Gospel Soul] [Tempo: 72 BPM] [Key: F Major] [Time Signature: 4/4] [Duration: ~3:45] [Primary Emotion: Peace, Hope] [Mood: Uplifting, Reflective, Reverent] [Theme: Redemption through grace and the healing power of mercy] [Intro]

[fast garden]

I want to make a song with two voices singing, each taking turns singing certain lines. Is there a way I can control who sings amoungst a number of voices?

[polar spade]

At lease add a space.. or two.. before [intro]. And you could try [interlude] instead as Intro can be seen as a parameter and just ignored, than a segment.

[graceful condor]

Yeah I did just didn't add in with my comment

[polar spade]

got it. Sorry. Just pointing out the obvious. 👍

[polar spade]

is this just an instrumental? or with lyrics?

[graceful condor]

it has lyrics Ooh… morning light shining through… (Let it rise…) [PAUSE]

[polar spade]

I used your parameters along with some random lyrics... used INTRO on one and INTERLUDE on the other... got a 5 and 15 second introduction on both gens.

again, it doesnt show so I'll ask, do you reference [verse] before the start? that may also distinguish from [intro] as if not, it will sing the lyrics as the intro.

[graceful condor]

here https://suno.com/s/5lgoq7FksdCre2ll

[polar spade]

Oh... you WANT it to sing that line as the intro? I see.

[graceful condor]

I see I forgot that

[polar spade]

yea. Id agree, it may be hallucinating to try to meld the intro line and the hook at the same time.

[pseudo blade]

note that hooks dont get used much anymore <---

check my workflow @tribal remnant

[polar spade]

when I ran your initial short parameters, it sang the intro fine and then my random lyrics.

When i ran your long parameters, it garbled the Hook line three times on one, then started singing verse one with no Intro line... and the second one, sang the Hook line and then hallucinated the rest of the song with no Intro line.

(I'll delete the songs as a courtesy)

[pseudo blade]

Hook is similar to Prelude if you think about it.

The Genius Lyrics Community Wrote This:
Genius no longer uses the [Hook] tag. If a section is labeled as [Hook], edit the lyrics or submit a lyric proposal to change it to [Chorus] or [Refrain]. If you want to put your lyrics up on Genius, don't use a hook.

[polar spade]

In this case, it may be treating the "hook" as a producer tag, anyway.

[pseudo blade]

Hook could be use yes, but I think it would fit better as an interlude/pre-lude depends on the genre.

[graceful condor]

I can just remove the Hook/Tagline

[polar spade]

yea. As was discussed this weekend... no one can define what a hook is. So, Suno doesnt know either. Call it something else. Let me try something I do... hold on.

[pseudo blade]

no one can ever define hook, until apperantly genius removed it in 2021.

[polar spade]

".....Let me try something I do... hold on....." OK... nevermind. it didnt work... it just sang the hook 15 times... lol

[pseudo blade]

wow.

@polar spade this is the english headers from what I know:
[English]
Intro
Verse (can be enumerated)
Chorus/Refrain
Pre-Chorus
Post-Chorus
Bridge/Breakdown (Mostly used in metal)
Interlude/Segue/Part/Skit
Instrumental -> Break/Intro/Outro
Solo (add specific instrument before solo)
Outro/Non-Lyrical Vocals
Build (EDM only)/Drop (EDM only)

[polar spade]

This is just a tool, and nothing more. sometimes it works for inserting as a producer tag style vibe when I'm looking for that:

I just put the "hook/tag" in my style, in quotes, in the prompt window. For your song, including the full parameters, only once it sang the hook clearly, then the intro clearly, and then my lyrics. But the other times, it just repeated the hook over and over. So, occasionally you may experiment with that along with your other bag of tricks.

[graceful condor]

Thanks for your help greatly appreciate it.

[polar spade]

Hope something helps. I'll delete all the sample songs made.

[pseudo blade]

im working on a language header guide google document and then @polar spade i'll send it to u if u need reference

[polar spade]

Sounds like a good thing. Just post it up in the workflow for everyone. That would be best. Honestly I thought there was something similar already there, but you can check if you wish.

[pseudo blade]

I did, but im making an enhanced ver.

[polar spade]

yea... for different languages? Thatd be cool. I'll check the workflow when its posted.

[pseudo blade]

its gonna take a while but it'll be worth it

[echo prawn]

I have marked the songs that generate female voices, is this AI stupid? The songs generated contain male voices

😅

[lost holly]

Suno knows vocal, female vocal, better then voice. Also sometimes Suno switches it up, but mostly that works fine. Unless you mix it up to much with conflicting styles.

[echo prawn]

But in my case, even if I specified it, it still generated music with male voices for me, which made me angry.

[lost holly]

Can you share an example, maybe i can point you out what the issue might be.

[spice birch]

When replacing or after replacing lyrics, when it gives you the alternates to review before commit" there are NO sounds on them. I can't hear what it sounds like before replacing. Anyone having this problem?

OMG! This edit area
It's completely messed up. It used to be pretty simple to replace lyrics. Now, you don't even know what's going on

[polar spade]

Can you do edits using the Legacy version instead? The older UI 2 is linked at the upper left of UI3 page.

[spice birch]

I will look for that ty

[spice birch]

Thank you. Got it to work

[spice birch]

Got another question. Let last few songs I have made, most of them at the beginning of each song sounds like a car driving by before it starts. Are you hearing this?

[humble verge]

i want a refund for getting 67 useless "extensions" lasting little more than 15 goddamn seconds. Why can't this idiotic thing understand something simple like EXTENDING THE ENDING OF A SONG?????

[raven turtle]

Has anyone any tips regarding applying Personas in 4.5 as it seems to drag in the riffs, Melodies, vocal melody and more When applying to a new song. V4 used the “spirit” of the source persona (as intended) but having trouble with 4.5 applications of it.

[glacial maple]

Does anyone have the technique for cover, I want to reproduce the sound I have but in better quality, because unfortunately the only good music I managed to get was of poor quality, now I try all the settings with "cover" but nothing works, the sound is permanently modified etc.

[sleek stump]

You might want to do remaster, it doesnt change the song, just tries to clean it up... tries... its not always successful, sometims you need to remaster in 4 and then back to 4.5

[glacial maple]

already tried....

okay i'm going to try it, thanks 🙂

[near flume]

are you putting "Male Vocals, Male Voice" in the Exclude Styles box?

[near flume]

How far before the very end are you starting the Extension?

[dreamy temple]

hi how do i get sound effects inb my song

[echo prawn]

I don't understand why Suno can't just do the exact calculations when generating a song.

He can't recognize the pauses in the lyrics very well.

[humble verge]

varies but enough that it should understand what it's extending. 20+ seconds.

[near flume]

post production, 99% of the time Suno doesn't do even the simple effects.

[near flume]

What do you mean by "exact calculations"?
What are you using as pause identifiers?

[gritty pilot]

How do you get suno to sing in a choir? I’ve tried [choir] and [multiple singers] and in the styles I’ve tried similar stuff, but it just doesn’t seem to work.

[polar spade]

Styles such as gospel, oratorio, requiem, shape note, chant, Gregory chant. Almost always if I use "boys choirs" I'll get a solemn high pitched choir singing. And of course, can try "girls choir" if you wish.

Your genre or styles may alter how much you'll get a choir as well, but we assume youre not going for a deathmetal choir, for instance.

Nowadays you can type in choir and click the "magic wand" and let Suno describe choir type styling or see if it does and edit from that.

[polar spade]

First, we can not yet place a general sound fx in a certain spot. Only a few, like [beep] , may show up around where you place it, though it may show up elsewhere as well and relative to your theme and lyrics. Others , like [applause], [clapping], [giggle] and laughing are more human created and can be summoned a bit more easily.

You may even try a sound effect with 3.5 or 4.0 as they seem better for inducing a sound effect with those.

[polar spade]

Covers should make a decent copy of a song. The song you're covering is the Audio. So you can slide Audio to 99%-100% to have the best influence of the original. Weirdness to 0%-1%. Style slider could be about 45%-55%, or possibly 0%, but lyrics are considered Style, so you be the judge if it needs it or not if you already have Audio.

I've had some good results to get the vocals to sound close to the original as well with :

Weirdness 0%
Style 45%-55%
Audio. 50%-70%

I can not address any "poor quality" . Covers i do seem
To do well in the fidelity of the outputs.

[gritty pilot]

Awesome thanks

[polar spade]

Just to note: Personas are more considered styles than Audio (though i may argue a bit on that). Nonetheless, throw in whatever style prompts you want in the style prompt box, but only set style slider moderately to NOT let persona influence as much. Similar to my previous post, I think (i always forget) that:

Weirdness 0%
Style 45%-55%
Audio. 50%-70%

Is enough to get the vocals in the persona's and the vibe, without all the riffs and melodies.

Try at least and my apologies if it does nothing better.

The other version is that both styles and audio are cranked near 100%, as we have , theoretically 2 of the three (original audio and lyrics) weighted over just 1, the Persona.

CHANGING LYRICS within a Cover session can makes for some garbled lyrics sometimes as Suno then has original lyrics, persona lyrics and the new lyrics to work out.

And pro tip: when creating the persona, only use about 1 minute of the original as the seed. Crop the best area for the vocal and vibe of the original song. This will help use less "seed buffer" to take up precious output area (length of song) and apply more Cover assets in your Cover, also typically giving you longer outputs as well, though Covers tend do decently at a consistent 3-5 minute songs anyway.

[viral pond]

Wonder why Suno do not stem all songs before putting it all together, instead of requesting stems afterwards, that way you remove allot of artifacts

[raven turtle]

You are a star! Thanks so much! I’ve been just creating a persona from about a 2min window generation. I’ll give these tips a go as 4.5 personas have been driving me crazy. Thank you again 🙏

[polar spade]

Excessive server load. Good idea, but the charge would be about 80 credits then for one generation (that would be two outputs , thus 12 different stems [doubled]. 😃 probably then Pro accounts would be 10,000 credits for $40 and premium 30,000 for $70. There would be no more Free tier. 😅

[viral pond]

Aww well it could been an option if you really wanted tho

[polar spade]

We cant prompt Suno to just give us a drum track or 3 minutes of bass, so it still seems to rely on composing
a "genre" based on its training. So thats evidence it would be a whole new model to make initial stems.

[polar spade]

Yea. Great idea. That then would take the place of a Recording Studio. Wipe out the whole industry. Or pump it up, one or the other. Its hard to know exactly.

[viral pond]

I wish stems and edits would not mic together with everything in your library but a separate one

[polar spade]

Yea. Dont thank me yet. And dont shoot me it it fails. 😆

[viral pond]

Hope for a better structure in the future for better library with folder and stuff, and random play recording edits and other

Workspace with nested folders, and each song connected with same prompt

Wishful thinking

[polar spade]

The browser desktop versions of PC/phone has a shuffle toggle button found in the playback banner, bottom of your page, if that helps. Im not sure thats what you mean.

[river zephyr]

Why does this not work

[small bridge]

did they lower the free credits to 40 per day or is mine being a butt?

[polar spade]

In some ways, as a duet is found in every genre, its not a genre and its not quite a style, similar to the idea we cant get Suno to alternately play a tenor sax and a bass sax in the same output. Putting it as a styles prompt is probably less significant. So it seems to take a combination of things to help nudge a duet.

[polar spade]

Should be 50. Yours is acting up.

[river zephyr]

Thanks so much for helping me out also it’s 50 unless you used one

[small bridge]

of course it is, three days now it's been 40 lol

[polar spade]

You use it down to 0 and the next day its only 40? Mention in bugs channel.

[timid lake]

Hi everyone, I was looking for a ‘convert voice’ feature to replace the vocals in a song. I like everything except the vocals. I tried the cover feature but it always changes the song. Anyone had success replacing/converting/swapping vocals? If so, could you share how? Cheers

[abstract flower]

Use the new editor and replace sections. Change the description to explain the new vocals. Generally, the section replacement doesn't keep the old accompaniment exactly, but stays close enough. Be aware that the length of generated sections won't match the selection, and will sometimes leave an extra tail re-playing the following section an extra time. If this happens, use the Slice command to cut the extra (you'll split the section into two, so you'll get one more), and delete the newly added section with the garbage tail right away. Zoom in well enough to align the cut; generally, you'll cut at the whole number displayed on top ("bar", although the unit is smaller than a bar). If you zoom in more, you'll see 4 more subdivisions of the whole. I didn't use them, but you may need to.

If you're going for stems + DAW path, expect the DAW be able to change the length of the vocal clip w/o changing the pitch.

To save on credits, try your luck by replacing a few consecutive sections at once. This has a drawback that Suno will only match the ends of the regenerated section with the unselected material. That is, if you replace verse-chorus-verse, the chorus may be regenerated with both the melody and accompaniment unrelated to the original. But e.g. verse-chorus is more reliable, as the new chorus will flow into the following verse seamlessly. But it's a hit-or-miss in any case.

There are other AI programs on the internet that can replace voices without changing the melody (and lyrics, of course). Cannot recommend any, never used them. Experiment.

UPD: Copy the new style description, as you may need to paste it again.

[timid lake]

Thanks @abstract flower for the detailed explanation

[abstract flower]

A question about the new editor: When I upload and then edit a song that I know has a stable tempo, at some random moments the tempo falls in half. It's not the constant regular floating of the tempo, e.g., 146, 155, 132, 142,.... These average to the song's tempo in the end, I believe. What I'm seeing is, at a certain place in the uploaded song (and all its future edits) the tempo falls down exactly in half: it jiggles around 146, then suddenly drops and jiggles around 73, like, say, 68, 74, 73, 69... Then it suddenly jumps upto the previous average of 146, again at a certain bar.

If I fix tempo to 146, the song plays fine to the point of tempo drop, then plays twice as fast, then returns to normal at the same bar number where the tempo returns from its half value 73 to normal 146.

Re-upload doesn't change anything at all, even the exact bar numbers where the tempo drops and rises are the same.

I don't worry too much about it, although bar lengths of sections are indeed weird, and the affected chorus is visibly shorter in the editor timeline. Since I only edit the song for generate the vocal stem and replace it in a Cubase project, it's tempting to fix the tempo, then randomise it a bit after the lyrics align with the instruments (not as wildly, usually ±1 BPM gives it a more natural feel). So I'd fix the bars where the tempo is half the normal tempo, if anybody knows how to do it. Thanks in advance.

[knotty stag]

I tried to do a few extensions recently, 2 seconds of silence, 4 seconds of sounds, not as good as it used to be to do extensions lately.

[knotty stag]

For one song, I did not ask for applause, but Suno added it anyway. Not sure if Suno just thought my song was so great, or what. I liked it.

[sleek stump]

I'm assuming its 4.5 you are getting short extends. 4.5 is not like previous versions, you cant just select the end of a song and hit extend. That worked great in 3.5/4, but in 4.5 we actually have to put some effort into it. Start by selecting back into the previous stanza, delete everything in the lyrics window except the parts of the verse or chorus you are overlapping. Then add additional sections with meta tags and lyrics, and then re-add your outro and ending.
This process will generally result in something a lot more substantial. Just giving 4.5 nothign to work with will result in 2-15 seconds of nothing or just garbage.

[polar spade]

If suno trained on publicly available songs as data, a few public song are live recordings. So a small % of outputs may naturally have an audience. Thats my presumption.

[sleek stump]

I can actually prompt for a live audience and get it in the intro and abotu half way through teh song, but 4.5 never seems to naturally give me an audience at the end 😛

[polar spade]

There are no specific copyright* restraints. Suno attempts to censor ALL publicly available content.

• It will give a "copyright" notice regardless of content it is censoring.

[stray flare]

hello everyone can some tell me how to make it so theres different voices for different people in songs cause chat gpt messing it up and not working right

[lost holly]

There is no guaranteed way.
What you can try:

1. But the lyrics you want sung with a different voice in brackets (), "", {} sometimes works as well.
2. Put Infront of each line [Vocal 1] and [Vocal 2]
3. Describe it in the styles that you want switching vocals.
   It all depends a little on what you want to achieve. Like each verse, each line different and so on.

[stray flare]

this is how its given to me but dont work

lol that didnt work

[polar spade]

Basic-intermediate level] - A simple means of, at minimum, changing voice is to extend on each verse and add a personas of your chosen vocal at the same time. Some additional editing may be needed afterwards.

[stray flare]

its carzy cause sometimes i get songs that are like duets but i cant seem to prompt it maybe a future update

[polar spade]

Use the Remix panel.
Choose extend. Pick a timemark with some audio.
Add persona that youve previously made (or someone elses)
Set Styles , ...meh... 78%
Set Audio slider about 85%
Weirdness 0% or your choice.
Click Remix.
Rinse and repeat.

[stray flare]

ok ill mess with that a bit

[polar spade]

I need to make a short, faster splicing than this one, but I sampled up this song on Friday. So here an example. A rough worktape, needs editing. Did in 30 minutes. https://suno.com/s/tA2ViUqxKHVv82pX

[jovial widget]

Why does the replace lyrics in the editor never work?

[polar spade]

Aligning the lyrics as precisely possible first then using the "Smart" blender option can help for changing lines or a few words, and especially a whole verse.

I often highlight the whole verse in order to change one word, this allows much more flexibility to make sure all words fit in the same space.

[jovial widget]

Tried that it just keeps repeating the old chorus and changes nothing. I can't even change the first word. For a paid feature it's terrible.

Brilliant for moving section around but changing lyrics nah

[radiant lake]

Hi! Sorry to interrupt your discussion, but I have a problem downloading the wav file, it's been like this since last week when the wav file was improved Pro. Not all songs, but so far I have 9 songs without the wav file. I don't know if everyone has this problem.

[humble verge]

It's useless. Doesn't matter if it's a song or instrumental. Quite often the "extensions" actually shorten the song, but most often it just replaces the bit that was selected. Maybe 1 out of 15 attempts actually does what is asked. I literally went through 67 tries with one song without getting a single actual extension. Not one! Just replaced endings. 670 credits down the toilet.

[polar spade]

I guess i got the gold-plated version. Sorry for the misfortune on your end then. Could be other factors involved. Hope you find the work around that works best for you, then.

[jovial widget]

3 just gotta hope v5 fixes it

[polar spade]

I have not seen anyone yet suggest, but extend off of one of your 15 second extensions, either from the end, from about 5 seconds in, from a decent audio point or do a pivot from the 00:01 point. What happens then? (Suggesting only things that work for me to get past the "same short" extensions)

[humble verge]

I select what I want extended (usually the last 15-20 seconds of a song) and write a prompt for it. Explaining, for example, that I want a low-register baritone blues guitar solo focused on flat wound strings with a dark, moody vibe.

What I have gotten, now almost 70 times in a row is not an extension but a replacement or a shortening of the ending. Several times just the instruments fading out. Not one single time - not one - did I get what I asked for.

[humble verge]

Doesn't change anything. It just makes a new short ending.

[polar spade]

Check. Thanks for trying.

[humble verge]

It's seems like a matter of random luck when it actually succeeds. A good 90% of the time, it doesn't. I've wasted so many credits on nothing. At least 1500-2000 this week alone on getting zero useable results.

[polar spade]

Option 2: And from one of your extension, you Get Whole Song, then Extend, so you are now including the previous (15 second) extension but reaching 2 minutes back onto the whole song.

Keep in mind im just working on getting it past the stuck 2-15 mark, not interested in also giving exactly the vibe. One step at a time.

[polar spade]

To be honest, with some combinations of prompts, 10% success in the same direction might be the maximum with 4.5. Anecdotally, 4.0 and even 3.5 follow prompts a bit better in certain genre styles.

[humble verge]

I really wish SUNO would just adopt a Midjourney type flat subscription instead of this credit based one. I'm pretty sure the vast majority of credits I've spent on this were spent on buggy and/or useless results. I don't really feel I'm getting my money's worth here. But that really goes for all AI-based platforms. Credits just isn't a fair way to pay for something that still isn't all there yet because you end up spending so much of them paying for the AI's errors and shortcomings. Midjourney at least factors this into their plans.

[polar spade]

Thats something to consider. $10 for 3 hours fast hour usage, the remaining, throttled back.

That'd definitely stop me from helping anyone, 🥴 , but I doubt they'd care.

[polar spade]

Throwing this out there as if credits were free: there's the option as I mentioned above. #📚┃suno-school message To make a persona in the style youre prompting for originally. Then extend and add personas in the same mix. Again, works great when I do it. And the example result of a folk song with 6 personas (vocal and style changes) on the same song melody and motif.

[humble verge]

I bought a lot of extra credits last month to finish a project, but I'm done with that now. It's just money down the drain most of the time. With the credit system, I'd be lucky to get a couple of songs worth keeping by the end of each month.

[humble verge]

Have you considered writing a guide?

[polar spade]

Considered doing videos. My real life gets in the way.

Another issue seems to be, what work for me, even if I video it, doesnt work for others as their prompts, lyrics, arrangement, personas, all are different. Nothings consistent. So there can be a 60% variability. So id mostly just piss people off more.

[humble verge]

But still it's nice to know what options there are - and some troubleshooting.

[polar spade]

Its all about throwing similar looking noodles against a wall and seeing which one sticks.

[polar spade]

Yes. Thats an encouraging thought. Thanks for the kind words.

[humble verge]

You give a lot of advice here, but you know how it is. The chat moves on, your advice gets lost. So if/when you have time, some workflow guides would be cool. With the understanding that what works for some doesn't work for all, given the nature of AI and individual prompting etc etc.

[half dragon]

Dang it. Been throwing spaghetti instead. 🙂

[polar spade]

But the sause stains are gnarly.

[sleek stump]

As far as extensions go, its mostly a process problem for the vast majority of people using 4.5. I have no issues getting a 1.5 minute extension in 4.5. its all about where you start the window, and what you add ion the lyrics window for content. I have a 90% success rate.

[half dragon]

Yep. Lots of cleanup after.

[polar spade]

🎶 Leave it like it is.
Never mind the turpentine, just...
Leave it like it is... its fine 🎶 -- David Wilcox

[polar spade]

A hair blower, a few brush strokes. One happy accident landscape, Bob Ross style.

[sleek stump]

I have a 100% rate on instrumentals, I use symbols and punctuation as things for Suno to work with and it does a really good job at it.

[pseudo blade]

I see a lot of movement here, I'm down to help here for a little. 😄

[polar spade]

Our pinch-hitter is here!

[humble verge]

My issue is with songs and instrumentals both, so the lyrics window doesn't really make a difference in my case. I very rarely extend from within vocals because transitions tend to suffer. As for how far back into a song I extend from - I think I've tried just about anything at this point. So I don't know what sort of magic you're working. The only times I've had any luck with extensions is when telling it that it's progressive rock, because progressive rock is known for long guitar solos. But it doesn't really work well to add a progressive rock solo to a swampy blues song or folk ballad.

[pseudo blade]

👌 I can help very well today

[quick wedge]

I can help too if it’s needed in here

[polar spade]

<---- Slowly sneaks off back to work.

[pseudo blade]

Note to Everyone -> Hooks don't get used anymore according to Genius: "Genius no longer uses the [Hook] tag. If a section is labeled as [Hook], edit the lyrics or submit a lyric proposal to change it to [Chorus] or [Refrain]."

So if you want to put your Genius.com, don't use [Hook].

[sleek stump]

I primarily extend from in the middle of lyrics because it seems to force the render to maintain the style, volume, and vocal set. Like I said, my success rate is 90% with getting good extends in 4.5. Learning how to talk to 4.5 hasn’t been cheap, I alone have spent almost 4k credits experimenting and learning, others have spent a lot more.

[north sleet]

What resolution does everyone make their custom cover art at?

[pseudo blade]

canva's 9:16 ratio works well

[sleek stump]

9:16 750px height

[humble verge]

That's nothing compared to what I have spent. And I definitely can't say I have anything remotely close to 90% success rate. I can't even comprehend that being a possibility.

[north sleet]

I have found it is all about what you say in the style to direct the AI.

[glass drift]

So, I was reading the chat before asking, but it seems like I got a good bug... I made a persona a few days ago and managed to use it to keep only the same voice in every new song I generate. But for what I read, this wasn't supposed to happen? Here's the thing: I made a new persona today and using it makes Suno generate basically the same song everytime (which it looks like to be the correct thing while using a persona?)

I'm so confused right now...

[humble verge]

I've tried so many prompts to crack that code. It doesn't seem to understand the concept of a guitar solo. At least not when it isn't rock n' roll. And even then it's like a 20% success rate.

[north sleet]

I got a wailing harmonica once that I keep chasing.

[pseudo blade]

#archived-ticket-help

[glass drift]

Came from there. Someone told me to ask here. Now I'm even more confused...

[pseudo blade]

okay let me see if i can help

U want to create a persona right?

I think depends on the prompt you made (genres/styles/tempo) it affects it but idk.

[glass drift]

I created a persona aiming to maintain voice consistence between songs. I thought it would work because I've done it before and it did. But the new persona makes Suno copy the song it was based on.

[polar spade]

Need-help is only for billing /accounts issues, according to pinned description. But many would not know.

Tech-prompt-talk is more a community of users helping each other out.

[north sleet]

@humble verge Have you tried in the lyrics [Instrumental Solo] or [Instrumental: guitar virtuoso]?

[raven turtle]

I had the same problem with v4.5. I took some great advice from Dasvon (big thanks) and cropped the source to only the verse and chorus which gave more of a flavour than the full thing, and it worked far better

[glass drift]

Oh, I see. Giving it a smaller sample so it doesn't copy the entire song. I'll try that

Thanks

[raven turtle]

Yep! I was constantly getting the melody riff intro and stuff copying over. A smaller sample totally fixed it

[humble verge]

In the lyrics? Wouldn't that just add a singer, singing it out? I put it in styles.

[sleek stump]

YOu need to use meta tags to direct specific instrumentation that you describe in the styles window. There are many examples of this all over Suno, most people dont clean up their lyrics before they publish. I would giv e you examples but I am currently at work...

[humble verge]

I'd appreciate examples if/when you have time.

[sleek stump]

Well, I have a track I havent cleaned out yet, check the styles for this track and then the lyrics and you should understand what I am talking about.... https://suno.com/s/LkeUGigw4Tsokoga

[humble verge]

so you direct it via putting commands in [brackets] - is that what you mean?

[humble verge]

[Outro - Slow riff fades, raw vocals]

[sleek stump]

Yes, those are the meta tags, Suno uses those as direction

There are varying was Suno understands meta tags.

[humble verge]

Ah, never tried that. Is it the same for instrumentals? Like, what do you do there when you want to extend the ending of an instrumental? Do you just place the prompt between brackets in Styles?

Or in lyrics?

[sleek stump]

They are not 100%, Suno can and does ignore tags, its part of the RNG of the generative models.

In lyrics I direct sections, I also use symbols and punctuation in place of words to make suno process them into music.]

I also NEVER use the Instrumental check box, I need the lyrics box to direct the instrumental for breaks, solos, chorus, bridge and outro, which are all still parts of instrumentals.

Let me get you an example of one of my instrumentals..

Heres one from my Jazz album, https://suno.com/s/WmlEtyM58IXMkPEL

[humble verge]

what on Earth am I looking at? That's just code.

When I want to extend the ending on an instrumental, I select the last 15-20 seconds or so. Usually I then type out what I want in Styles. Like: Hard blues with a low-register, deep baritone guitar solo in the lower octave, warm and growling.

and SUNO just ignores that completely.

How do you come up with code like that? What does it even mean?

[lost holly]

Its kind of an attempt to provide suno the wanted melody, nothing more. Iz can be fun to find out what you can do with prompts with suno 😄

[humble verge]

Yeah but you don't just type code like that out of your noggin'. That's some secret knowledge.

[lost holly]

The letters are notes, like when you play the guitar or a piano. The other symbols are similar to sheet music. No secret here^^

[humble verge]

Uh huh.

Yup yup.

[lost holly]

All typical notes C,D,E,F,G,A,H,C

[humble verge]

all typical sheet music: ., - -- ^ . - `.. , | *. - .. -

man if only i had known, I could have done extensions so easy. Here's my blues solo: . - .. | .- .. - ., . _ - . .

[lost holly]

If it works 😄 who cares 😄

[humble verge]

I'm fine with it working, but I don't believe for a second that's typed out

[lost holly]

But yeah, like ginger said, try adding the styles in the lyrics section like [Blues guitar solo] or something similar, might work

[humble verge]

I shall try that. I just thought whatever went in the lyrics section would be sung.

[lost holly]

Sometimes it is sung, sometimes stuff from the style section is sung 😄 there is much we dont know for sure on how it works with suno

Either way, good luck 🙂

[sleek stump]

Those are in fact typed out lol... I dont use AI to write my music 😄

[vivid bough]

also huge thaanks to the devs and the owner of suno for letting us make fire rap beats and more :)...

[sleek stump]

It doesnt follow it, Suno mostly ignores notation currently.. but.. it does use it in a sense, same as symbols and punctuation, it will add to the arrangement, and length. A few of us figured this out pretty early into 4.5.

Its also why I can get full instrumentals without the switch being on

[north sleet]

Punctuation in the lyrics will add pauses, delays, and the little nuances when sung.

[sleek stump]

Pretty much, from what we have found...

Mind blowing yes..

Suno isnt trained on Notation, so it doesnt know what to do with it...but it will use it to do somethihng.

[lunar anvil]

Were you able to get a guitar solo?

[sleek stump]

In older models we had varying degrees of control with tags, 4.5 opened up the gates

3.5 was very limited in what we could use for tags... and there were some real big issues with its output... shimmer killed a lot of good songs

3.5 though was able to do some things 4 and 4.5 can not... live audience tracks are awesome in 3.5

[lunar anvil]

I am a heavy tag composer. Here is an example of tags I used in V4.

[Chorus]
[Powerful layered vocals over rising timpani and strings]
[Vocal harmonies]
The echoes still remain,
Our fire burns untamed![Brass hits]
We do not break![Unison Hits from full Orchestra][Drop]
We are not bound!
[Sforzando Hit]
Our strength eternally profound!

Here is the track We Are What Remains https://suno.com/song/1d7995a3-d5cc-452e-8fad-aefb1e6e759b

[sleek stump]

Dont confuse the render model with the editor features...

[lunar anvil]

I rarely use Cover or the editors. The Cover feature was broken for so long that I never used it and now it is a mystery to me.

[sleek stump]

Cover is much better now with the sliders exposed...

[lunar anvil]

If it doesn't know how to interpret the tag it hallucinates in its place. (edit: fix spelling error)

[polar spade]

Interrupt = Interpret? (Maybe a misspell?)

[north sleet]

Misspell.

If I was at work I would edit to fix it.

[frail badger]

came here to see if there is an api #featurerequest

[sleek stump]

There will never be an API, SUno's founder has made that clear in a few interviews.

[frail badger]

I'm a business willing to plonk down millions of Vietnamese Dong for a specific use case

[polar spade]

2.4 million VND = $100 USD

[river zephyr]

https://suno.com/s/fb9v75pv3BMkYPSz
Why does it cut off i tried extending it still cuts off last bit

https://suno.com/s/G41vqmHM9uRWLvSG

[stuck dust]

If you are extending the song, only include the lyrics you want to include in the new section. You copied and pasted the full lyrics again.

For example, if you want to extend after the chorus, delete everything except what comes after the chorus.

[river zephyr]

Thanks that’s so helpful

[stuck dust]

Also when you get the extension you want, you will have to select "get full song" on the extension track to combine the extension with the original. Doing this does not cost any credits.

[river zephyr]

Thanks for all the tips Yeah i could do that or I have other tools that works to combine and edit

[noble raven]

How the FUDGE do I make a duet for male and female everything I've tried they were singing it female but the line is male

[peak veldt]

Duets are a mystery - *I spent so many credits trying to get * a good Tongue-Twister song to work with two parts singing, (Male & Female), one was the accomplished twister singer trying to help the other get through it, and to no avail, it messed up the parts... REPEATEDLY. No amount of tags within the song could get it straight. I gave up.

So I haven't found it yet, not even in v4.5 which I was hopeful.

[noble raven]

I'm now seriously lost the ai is singing the female part with male voice and vise versa for the male

[chilly wyvern]

You can try to make your point by putting (* song has two singers, man and woman ) ( man singing ) ( woman singing *) to lyrics and getting lucky with generation. Some genres seems pretty much impossible to get it right. Example of some kind https://suno.com/song/1cbbf1eb-3656-45dd-ac5f-f97465a23585

[noble raven]

I did that for my lyrics like male singer only

[noble raven]

Then the female just sings that part

[peak veldt]

Keyword: Lucky Generation, is right.

[noble raven]

I've tried alot of variation got months nothing

[peak veldt]

We've all felt the pain. I commiserate.

[peak veldt]

By the way, Rand - beautiful song.

[noble raven]

The heck suno won't generate the song again it says I'm out of credits even tho I have 10 more

[peak veldt]

Ugh - waiting for the next day is killer.

[noble raven]

What prompt did u use for this ?

The genre then duet?

Or did u genre
Hard rock, male and female duet ?

[chilly wyvern]

Let's see. This is remix of genres "bolero, ragtime, dusty, poetic" with 1% audio and 5% weirdness.

[rigid pasture]

Will the new policy coming from YouTube include monetization for the Suno Ai Song channel or will it be demonetized?

[stuck dust]

The new YouTube policy does not concern AI. It is a general change, it will actually affect videogame videos far more

[rigid pasture]

@stuck dust No, it says low effect channel, AI channel, will be demonetized, so will this have any impact on Suno AI song channel?

[stuck dust]

The actual documentation was misinterpreted by blog sites. AI specifically is not affected. The change is largely just to try to punish unoriginal/repost content.

[rigid pasture]

@stuck dust I use AI song on my channel and edit the clip of another video song and match it and upload it. Will there be any problems? I use a maximum 10s clip.

[rigid pasture]

Ok

@stuck dust I will send you the channel. You can watch it.

[polar spade]

I have, in the past, generated the output. Lets say its all male, so then use Replace and in New Lyrics adding [female vocal] before the same verse and putting "female vocal" in the style prompts. Then run some generations until I get what sounds like Female on that verse. Then move to the next one.

Legacy Ui might be easier for that, but your choice

[limpid fox]

Hi, I spent 1000 credits trying to create another version of the same melody using different Persona and lyrics, but I can't! The results are always bad.

[hard beacon]

and why didn't you use the remix function?

don't you / didn't you <-- ??? whatever the correct grammar is 😄

[limpid fox]

I will give a try.. thanks.

[hard beacon]

let us know if it works 😉

[hard beacon]

@limpid fox : any results till now?

[hushed mesa]

Hello! I created a song but can't download the WAV. I pay for Pro. I can download any other song in WAV but not this specific song. Is this a bug? Thanks!

[fresh wave]

Does suno have a way to use your own vocals and generate a backing track off of it? I asked in the main channel and someone got mad

[hard beacon]

schmitty: take a look into #1372994765196361774 .. there is a post about the problem

[polar spade]

Yes. It appears to be a bug.

[hushed mesa]

Thanks! I'll be patient and see if the Suno devs can fix it 🙂

[polar spade]

Some have had success to copy the file and try downloading that new (copy of) output.

[limpid fox]

I got a very usable result using "cover" with "persona". Not as good as original, but usable. Thanks.

[hard beacon]

you are welcome 🙂

[knotty stag]

Could you create two versions, one male and one female? Then get stems to get vocals for each and instrumental. Then in a DAW take the three parts you want: music, male, female, and cut out the male and female parts as necessary. Bit of a nuisance to have to do, but as a last resort you could consider that.

[hushed mesa]

How do I copy?

[kind sierra]

I finally got a duet to work.https://suno.com/s/ndU7A4909ctIVwTu

[polar spade]

Go to Edit UI page. Highlight the whole song, click 3-dot menu under wavefile, click "Crop" , save new file. Find cropped new file in Library.

In Legacy, same. Select crop. Select the entire song. Click OK. Go to Library to fine new file.

[hushed mesa]

Thanks! 🙂

[polar spade]

Yay!

[regal bay]

good morning, just curious where I might be able to get support on a ticket I submitted 6d ago.

[pseudo blade]

#archived-ticket-help

[buoyant gale]

I just upgraded to Pro. My genre is instrumental fingerstyle guitar. 4.5 is good but the free 3.5 model was way better in creating melodies and harmony that's why I upgraded thinking 4.5 is better. Sorry to say this but the melodies are boring and not enjoyable in 4.5. The prompt in 3.5 was more accurate for me. I wish there is a refund though. I hope it gets better in the future because right now I am disappointed with 4.5 instrumental 😢

[neat nexus]

Have you taken advantage of covers and personas in v 4.5?

[ruby urchin]

Is there a „right“ syntax for v4.5? Sometimes the metatags are sung along

I use only the mobile custom textfield for the songtext. metatags included

[neat nexus]

That's just AI being AI.

I've had it happen to me too. Usually that just means that whatever you're trying to use for a [Metatag] isn't being recognized. Also check for instances with missing brackets like so: [Metatag which can cause the AI to think it needs to sing it.

But yeah, I had an instance recently where it was singing [Chicago Blues vibe] on me half the time. Had to listen carefully to figure out which tag was messed up and remove it. There's nothing wrong with trying stuff out just remember that not everything will work.

[ruby urchin]

Thanks for your answer. Then i have to try and look if somethings missing. 👍🏼

[buoyant gale]

Yes I did but the same unenjoyable melody. I already used up 400 credits today and the same result. I hope it gets better in the future.

[neat nexus]

I'd try shifting my prompting before giving up on v 4.5. Most of the time a new model will require a different type of prompting.

[buoyant gale]

I've tried it already. This is instrumentals, not songs. I would make a request to support that the music I like in free v3.5 be upgraded to Pro. There are 20 of them.

[glacial maple]

Hello everyone, I have a very important question; the AI ​​often produces sounds that are unusable in terms of quality, despite well-constructed prompts. I would like to know why some sounds weigh more than others in megabytes (this is due to the quality of the piece obviously) but why does the AI ​​make 1/100 a sound that is of excellent quality and most of the time degraded instruments? What are your methods to remedy this? Thank you very much ☺️

[sleek stump]

Depending on model version, it may be a genre specific issue. You did not say what model you are using. When 4.5beta was first dropped, a couple genre (mainly hard rock/Metal) had excellent vocals, but the entire band behind them sounded like they were playing from deep inside a mud puddle. This has been mostly remedied as they continue to work on the model. What model are you using and what genre and instruments are you getting? It would also help if you explained what you mean by "degraded".

[polar spade]

Just now, I took a 3.0 instrumental fusion jazz song from May 2024 (2 minute limit with a cut end).

https://suno.com/s/MEpa19Mnop9Gi1GM

Made a Persona. Covered same song to 4.5. So we have Audio showing at top and Persona attached.

Audio Slider to 100%. Style Slider to 100%. Weirdness to 0%

Click Remix. Done.

May work without a persona, but for taking 2 minutes, I dont mind. I can use the Persona and make a variation album of the same.

Results: 4.5 song. Same melody/instrumental. After 2 minute mark it extended it further for 5:20 https://suno.com/s/wTeEJAF2nZUbxrE3
the second output with a bit varied in chord progression and so a note shift once I think. But virtually 99% identical.
https://suno.com/s/JhgcyRN6Vq6v8jSd

So dont give up. Its easy peasy to do.

[pseudo blade]

wow.. you were able to send 3 links 👀

[oak mortar]

question, i wanted to add like guitar flicks after a certain word or what not...i swear i saw someone post something on here that gave me the idea...i tried it i may have don't it wrong..is there a certain way? EX: word word word [high pitched guitar flick]

[onyx pumice]

I accidentally deleted a song I did not want to.
Does Suno keep copies of the songs that I could maybe contact them and restore that song?

[sleek stump]

If you trashed it, go into your trashcan and recover it there. Its the little trshcan icon at the top right of your Library

If you did the actual delete from the trashcan, then nope, its gone gone.

[onyx pumice]

Well. I tried that. I was trying to clean things up and I deleted a large group of song that didn't come out right and I accidentally added a song I didn't want to delete. I didn't realize this until later downloading songs on to my laptop. So I did the full on delete.

I have all the data for the song. So if they have copies of them it would be easy to find.

[sleek stump]

If you did the hard delete, you will likely not be able to get it back. There is a clear warning as you delete tracks that they will be permanently gone.

The only thing you can do is email support and ask if it can be recovered. Beware the wait is long, weeks long, and you may get a hard nope as a reply when they get to you.

Or you could get lucky and they find it for you.

[onyx pumice]

Lucky, it's a song I wrote all the parts to so I have the original DAW recording of me playing everything and singing so I can recreate it as I had Suno do it the same way just more polished and a better voice then mine. Just sucks to have to do that over again when I have the data copied from the Suno version.

Thanks, I will email them. Maybe I'll get lucky.

I would think they have copies of the songs we make. How else could they keep track of things. It would be a free for all if not.

[thin flint]

Did the create UI change recently? Th first box I come to is for workspaces, not to add lyrics.

[faint meadow]

Having a tough time trying to insert male vocals into a song, Suno is consistently using female vocals no matter what I try. I'm using [brackets] and (parenthesis) for backup vocals, I'm being very descriptive, etc. Thanks for the help in advance.

[pseudo blade]

hey thanks for coming here!

Let me help you out

do [Male Vocals] / [Male Voice] in both styles box and lyrics @faint meadow

[lusty juniper]

unable to select a persona 😛 the window is bugged and stuck at the very bottom

[polar spade]

Is that the persona page or the drop down menu on the Create page? Well, either way, does the other one work?

[lusty juniper]

It's the drop down at the create page.

[polar spade]

Well, either way, does the either one work? Try the other one. 🙂

[polar spade]

A lot of different bits of Suno are laggy this week. So I feel for ya.

[daring lava]

S/O to @polar spade and @sleek stump for always being thoughtful and helping people learn and grow ❤️

[polar spade]

So, did the main persona page work?

[tribal rivet]

persona is too much on down on the site and i cant use it

[polar spade]

MTH just mentioned their drop down persona menu on the Create/Remix page was laggy or sticking. So, maybe something going on as far as being buggy.

Personally, the menu has always been iffy for me to always show the entire list. I have a habit of just using the main Persona page, and its easier on my eyes to find the right one. Once clicking on it, it still auto loads into the create/Remix page we are working on.

[tribal rivet]

it helped to shrink the browser tab

fix this....

[buoyant gale]

Thank you, I will try that

[polar spade]

Hope it works, my friend.

[maiden orchid]

Could anyone give me some advice on how to make the songs different when you use the same persona? I have this one AI voice I love but every song wants to start with the same music as the first song made?

[sleek stump]

Use the sliders, try Persona influence at 70, Weirdness at 75, and style at 80.

Then start turning down teh persona slider

[buoyant gale]

If I make a persona and cover a free plan song to Pro 4.5, does it change the license to commercial use?

[sleek stump]

No

You can take your prompts and lyrics and do a new generation with a persona under a subscription.

Simply doing a cover will not grant you the ownership

[buoyant gale]

Okay I will try that. I know this is a silly question but I just like to make sure, I'm using Pro and if I use the old version like 3.5 does it grant me ownership for using the old models?

[sleek stump]

ownership is all the same no matter what model you use.

[buoyant gale]

Thank you

[eager wolf]

hola buenas noches mi suno no funciona

que hago?

no me carga

[urban canyon]

i think you meant Audio influence slider? why would you start weirdness at 75...and also why start at 70 the audio influence.... i found the song only changes under 15%...thanks for explaining

[keen knoll]

Has Anyone had success making music similar to Peaceful Sleep - Nier Automata?

I've used Suno a lot and been having trouble hitting it, I've been gotten near it with Bagpipes and Chinese Instruments but I haven't been able to achieve it. (I've tried putting fantasy lyrics in as well but it's mainly the instrumental and style that's the hardest to hit.

https://www.youtube.com/watch?v=syIQMFx2yI0

[raven turtle]

Well it’s been a while but like a returning old friend I got some “Shimmer” today! Interesting….

[polar spade]

Nothing? Huh. Ok.

[keen knoll]

I just came to check it and the message was removed.

[polar spade]

It wasnt important. No worries.

[keen knoll]

This is the closest I've gotten. https://suno.com/song/d73196f4-7440-4161-a172-1c441a385d95?sh=EFSR2nEFmWxz1Hzc (And it's far off)

[keen knoll]

I was reading the first sentence and it was deleted. Didn't see the notification. I appreciate you taking the time to respond

[polar spade]

Ill post again. No worries. Everyone seems asleep. So i didn't want to just hang there. Lol.

[polar spade]

I have found it fruitless to try to help someone get a "sound like" an artist or another artist music without at least getting some basic prompts to start into the genre.

What i do, and likely best for you, is to use the many sites that offer an analysis of these and give back suggested prompts to fit the genre or vibe. Audioaidynamics is one of a few I use.

From there, you might then post your output and ask how to get a more specific sound.

Prompts like ethereal , whispery, and dreamy might help as well.

[keen knoll]

Was just curious if anyone else got something similar as I can get pretty close to many genres.

[keen knoll]

Thank you. I can try that. Main thing I need to learn is how to prompt 4.5, seems each update changes too. It doesn't seem to know the instrument used in the song so I've been trying to crack that nut

[polar spade]

Posting your own attempt helps others to guide you. So thats good you did.

Im heading back to sleep though. So , you might need to post later when a few more get online.

[keen knoll]

I can post in the morning / day too when more people are on to see if someone already had done it in the past and has recommendations.

[keen knoll]

No problem thanks for the response, great idea I'll do that tomorrow as well 🙂

[humble verge]

Nope. SUNO doesn't seem to understand that EXTENSION doesn't mean REDUCTION - or that it's not supposed to just replace the highlighted bit. I mean, it does understand it sometimes. In my current attempt, it understood the assignment exactly four out of 67 tries. None of which were usable and just added 10 seconds of silence to the ending of the extended bit. I ran out of credits trying this, so now I've given up on it for the time being. Waste of money. Maybe I'll give it another go a month from now, since daily credits don't stack on this platform.

[sullen granite]

hi my music distributor asking for documentation of songs i generated with suno ai
i have send invoice of my pro plan purchase and give terms and genral commercial license links of suno ai but they all reject and reply provide proof which shows you have control on song please dm me solutions

[polar spade]

If you wrote and produced at home, how would you show proof then? I guess you must have said it was partially produced with AI?

Who is this distributor? Maybe a use can say whats worked for them.

[sullen granite]

did 22 plus email converation with them but they didn't accept any proof distributar is toolost

[polar spade]

Apparently not a prominently known distributor? I want to read their TOS. Whoever it is doesnt have a legal department. Lol.

[sullen granite]

but same distributor release loudly ai music loudly proving certificate for each genrated song why not suno ai providing it ?

[polar spade]

You mentioned: Toolost

[sullen granite]

loudly is also genrate ai song

like suno ai

i distributed loudly songs generated with toolost too

[rain lance]

Is there any way to learn how to use the editor? Sometimes I can get it to work but othertimes not. My replacements are all over the place and they don't replace with the words I'm choosing. I'm also getting a lot that say aren't a part of the timeline and I need to drag them in, and I don't know why

[sullen granite]

suno should provide legal proof that song generated with suno ai which can be use legaly in future if anyone claim of song theming, music,beat and style

according to tos paid users = ownership, but how a user can proof it for a particular song ? user only have a invoice and nothing else, how can a user can proof of song generation with suno ai legally ?

[unborn star]

They should do the opposite, they should show, that you used their work, instead of just claiming it. But how is it, when you create a song by yourself? You don't have a license for it, because you created it yourself. How does this case work and isn't it similiar to AI Content?

[echo prawn]

To be honest, I am really not satisfied with the speed of Generate music now. Because it takes too short a time and produces half-finished products every time.

[sleek stump]

If you created the track in your own space using non AI tools, you would have no documents to show either. The only documents you would get would be if you registered your song with the Copyright Office. It is the same with a Suno track, sort of. You can register the human made parts, but anything that is generative can not be registered.

[echo prawn]

I would rather they generate a song with deep computation

I don't want to get a bunch of Scheisse

[sleek stump]

I'm looking at their TOU, having a hard time finding their policy on AI generated content...

I AM seeing a lot of the typical gotcha's that distributors are pulling these days. Industry attorneys would likely recommend not using.

Its also a European based business/distributor. But then they have an office in New York... thats a bit confusing

[polar spade]

Overall, not much of an answer can be found here. I usually steer clear of any company that doesnt have an established track record or one that I can not rationalize.

[sleek stump]

Ya I am going to just assume they are another AI-hostile platform, They only mention the term "documentation" around their TOU and FAQ pages, but they never detail what documents. The onus is still on the person to register the track though, as that is the best "legal" proof one has of ownership and control. If a track is 100% AI then that is not going to happen currently.

[buoyant gale]

What if I record my Suno music with my own instrument and made significant changes in the song, can I register it for copyright? Since it's entirely recorded and repurposed by me

[polar spade]

In all fairness, you should go to their website and click through the application and see what you think. Decide for yourself.

[graceful condor]

does cover shorten songs? like from 07:00 to 03:45.. and still not sing all the lyrics?

[pseudo blade]

Are you usin v4.5?

[graceful condor]

yes and it was my first time trying it out.

[pseudo blade]

ahh okay, can u send me what u mean?

[graceful condor]

I had some long , long drawn out words like "death" and "rise" and got no outro spoken word in the 07:00 one. i did cover of that one and it just missed instrument solos, chorus, final verse, and outro.

[pseudo blade]

okay, ahh yeah. well this happens if the spoken is too long

[graceful condor]

it's messed up because my first gen of this song everything was sung right but it just wasn't gospel enough.

[fair adder]

hello, i come because i'm confused about one point. what does the weirdness slider do? i can understand style influence and audio influence, but weirdness? i'm kinda scared to touch this one lol

right now i'm just trying to use an older song with braind new lyrics but keeping the structure and the vocal melody as much as possible

[buoyant gale]

There seems to be a problem in downloading wave, mp3, videos files. I know this has been tackled before but I don't have this issue in free plan

[polar spade]

It seems to do unusual things. Sometimes go off genre. Sometimes just combines non-congruous (nothings the same) noises of possible combinations. Often what we think of as hallucinations, weirdness will do much more of that. It tends to be more treble, more squeaky, more random.

Less weirdness and sometimes it will just change an instrument that may not be expected, or change up the chord progression out of key here and there.

On lucky times weirdness will make odd noises but some or much of it will be in rhythm, so its good enough to make a song. But when dropping the Audio and only running a Persona, Weirdness doesn have as much effect up to about 70%, so it becomes a gentle variable.

[fair adder]

m'kay so to summarize it better keep it off it seems

[polar spade]

Haha... I guess. I will just put about 20% sometimes and it may add a whistling person in the song. Its been more fun to experiment than to use specifically.

I think the one time I may like to use is if I am taking a 3.5 song, Covering to a 4.5 with a persona. So Audio is 100%, and Style is 100%, so it will be a perfect copy of the original 3.5 version. So maybe I want a bit of change, then I will slide weirdness up and see what happens.

[fair adder]

mmh okay i see.but i don't want a 100 recreation, even if it's indeed a v3.5 to v4.5. i just want to keep the same audio and structure and vocal melody, but i'm jsut changing the lyrics

but you're right i need to experiment more. the workspace is not very easy to handle with only a keyboard though lol

[polar spade]

Yes. You have more of a challenge. Honestly, I will they had ALT key shortcuts as well, instead of moving a mouse.

[fair adder]

plus i'm a screen reader user too loli mean sometimes in the workspace when i clic a song for listen for example, most of the time the focus moves me to anther song, generaly the most recent one

but that's another story

thanks for your answer about the weirdness

[brazen yew]

Tells suno to make the vocals a horrendous noise. Suno gives me Aretha Franklin...

[ivory thistle]

Yo guys, how does the new Song Editor work? I'm coming from version 3 and don't really understand it. Do I still work as in version 3, where I do every section seperate?
Currently I have my Intro and added a verse section. Now I want to add chorus stuff in a new section, but it isn't using the new lyrics. Do I have to extemd it? But then I can't add any new lyrics.

Quite confusing.

[rare oak]

Hey ! Someone has some tips to make Hardstyle/Rawstyle with the 4.5 ? it worked fine with the 3.5 and 4.0 but I just can't generate something good with the 4.5

[pseudo blade]

just try to remaster it!

[fading barn]

I cant generate Drift Phonk. Any one has had any luck?

[rare oak]

This doesn't fix my problem, I'm not gonna make hardstyle in v4 and each time remaster on v4.5

[pseudo blade]

well.. firstly pls dont be rude, im just a 15 yr old autistic kid.

[rare oak]

Wasn't trying to be rude sorry mate

[pseudo blade]

i personally do not have paid, so i cant help, plus I never heard of hardstyle.

[rare oak]

ohh okay

[keen knoll]

I asked this last night but most people were sleeping. Has Anyone had success making music similar to Peaceful Sleep - Nier Automata?

I've used Suno a lot and been having trouble hitting it, I've been gotten near it with Bagpipes and Chinese Instruments but I haven't been able to achieve it. (I've tried putting fantasy lyrics in as well but it's mainly the instrumental and style that's the hardest to hit. This is the closest I've gotten. https://suno.com/song/d73196f4-7440-4161-a172-1c441a385d95?sh=EFSR2nEFmWxz1Hzc (And it's far off) I can ask in general too but I felt tech-prompt-talk felt like a better fit

https://www.youtube.com/watch?v=syIQMFx2yI0

[sleek stump]

4.5 is not like previous models, you need to be specific and verbose in the style window with what you are looking for . Simply putting hardstyle or rawstyle will likely result in 4.5 not knowing what that is. You need to be specific with what styles you want and then include any sub styles... Then hit the magic wand button and let the AI do its enhancement. Additionally, you should reinforce your instruments and style types in the lyrics with meta tags.

[polar spade]

@keen knoll As I mentioned. Its nearly impossible to get the "sound" you want because Suno goes in different random directions depending on the styles you give it.

But here are three examples with the general prompts of: Atmospheric, Dreamy, Ethereal rock, neofolk, Electronic, experimental, psychedelic rock, irish folk, easy listening, pop funk soul, jazz, uilleann pipes

When you run this, you will likely not get anything like the examples. And mostly be disappointed. But I do like the beginning of my second one. Sounds like a nice sync for a movie. But I ran it with a persona to help on the vocal.

So you'll get some random combination within the genre of Neofolk and Psychedelic Rock. I added a quick 4th one to just prompt Psychedelic Rock and Neofolk, just to give an example. Even running through a genre analyzer , they sound quite similar to your attempt, id say.

https://suno.com/playlist/d20685dd-12c7-43d2-9e05-29c17073c5cb

[fathom cairn]

Question. I was poking around and was told that there's a big difference between ownership and copywrite and authorship. Was wondering if you've heard of anyone loopholing it by basically getting the music ownership first, then musicsheeting, then copywriting the musicsheet and not the audioversion that you already own. And basically finally be like "yea, i have copywrite on the blueprint AND own the audio so... ya" ?

[clever tinsel]

Does anyone know or have any idea why SUNO completely ignores or skips over entire verses!? It is soo frustrating, I can’t even edit the parts to get it to do what I want, I have wasted soo many credits already.

[near flume]

v4.5 understands genre tags better than 3.5/4 that's where the problem lies. Try to describe the sound you're looking for the best you can in the styles box, you have 1000 characters to work with.

[near flume]

check the structure elements in your lyrics, you could have an open tag or other anomaly in the tag or some special character in the lyrics. Suno typically doesn't skip verses.

[clever tinsel]

Thank you for the input. Would you mind being more specific or give me an example though, so that I know I am on the same page as you are describing please 🙏🏼?

[oak gazelle]

Any tips for making suno properly utilize male and female vocals? Just using [Male vocals]/[Female Vocals] doesn't seem to matter and Suno continues switching the roles at random. And yes, I've regenerated at least a dozen times and it still messes them all up. I even tried using the editor, but for some reason the Female Vocal lines don't even appear there.

[sullen granite]

i didn't get you ! what your meaning about the fard ? is suno ai doing any type of fraud ?

[polar spade]

Thanks. That was an old conversation. I've have since removed it. I've been a member for over a year with Suno. Obviously I dont think that, so you misunderstood something in the translation. Sorry, mate. Best of luck with your songs.

[polar spade]

I think the more exact (or complicated) you get in getting down (micro prompting) to sub-genre sounds, the more you have to listen, tweak, move a prompt, listen again, tweak, rearrange, add a prompt, tweak, etc until you get the sound you're looking for. So, it may be possible, but from experience some things can take dozens, if not more, attempts.

We could be better off to emulate the genre the best possible on an uploaded home recording or DAW creation we made and see how much Suno can pick up on it.

[little vessel]

anyone have any idea how to get suno 4.5 to stop rapping every other verse? Is this a known issue? Been using Suno for almost a year now and it wasn't really a problem in previous version but now it seems almost every song, it seems to want to rap or very quickly sing the lyrics, even when negative prompting rap or trying to specify slow singing? Anyone have any luck? It's almost every song in almost every genre I tried, burned through hundred of credits of bad generations with this annoying problem, finally at my wits end with it.

[keen knoll]

Thank you, you got much closer! I'll work with what you prompted and try to get even closer. You got a lot of the similar instruments and style!

[sleek stump]

This has 100% to do with a genre you are using in the Styles window. If you are using "any" type of rap, hiphop, triphop, or trap or other genre that generally contain rapping in your styles, you will get rapping. There are a few ways to try to tackle this.

1. Remove the styles that are causing the rapping
2. Add Rapping or other styles to the "Exclude Styles" window in Advanced Options (just above the sliders)
3. Use the sliders to vary the influence and the weirdness.

Sometimes a combo of these are needed.

[terse plover]

i have soon used up all my credit. i try to do a trance with real nice sound. But its hard. Has someone good promps for reel good trance music

[thorny cypress]

It could also have to do with how many lyrics you've got in that part. If there are a lot of lyrics with long lines, it may try to rap it to squish it into the available time.

You can try creative use of symbols like ... or -- in the lyrics to get it to draw things out or add space between lines. (Warning: I lack experience with newer features so unsure if this still works.)

[solar sequoia]

Has there been any word at all about making the app work better on Android?
Can't change my profile avatar (literally the first thing I tried to do when I paid for a subscription) I can't make personas (I have some template songs that work most of the time but I bet 7-10% of my credits get ate up from that) I don't have the "magic wand" for the style box everyone is talking about (which If I had one I could learn to communicate with the Ai much faster) I have no "Listens compared to Likes" display on my stuff even on the homepage. Which I have to ask... are all the times I've listened to my music counted on that? Because I'm rather new and am not exactly sure how that works. I don't want folks going "Oh it's only got 6 Likes and 120 listens" ( 🤷‍♂️ I write my own lyrics and I like hearing them) and my song gets overlooked. I'm sorry for writing a novel but I got questions and a lot of the time it's just blundering around trying to figure out all the nuances. Oh and I've emailed Suno support about my problems but from what I understand they are weeks behind.
Thanks in Advance for all your answers.

[sleek stump]

All I can offer is that Andriod app has lagged behind everything else since it was released. They are adding features to it, but its been slower than the iOS app, and far behind the desktop web app. The desktop has all current features, none of the mobile apps have all of them. For things like profile avatars and other features, currently you will need to use the web app in chrome browser. Also, the Suno home page and its trending stats are often broken or lagging in stat updates. I would not suggest putting too much stock into getting on the home page.

As far as the magic wand, that is a 4.5 only feature

[solar sequoia]

Thank You I appreciate your feedback. It's been a little frustrating... and my "Business Brain" starts asking questions about percentages and ratios.
But other than the complaints I have above I'm completely digging on this app.

[little vessel]

it is 100% not the genre I'm using, here's a couple style prompt examples, and yet it will most of the time just start rapping one or two of the verses after not doing it for others. "Cinematic, Epic Orchestral, Spoken Word Anthem, Inspirational, Heroic, -rap" "Classic Folk, Acoustic Americana, Narrative Ballad, Melancholy, Bittersweet, 3/4 Waltz, -rap" "Acoustic Ballad, Storytelling Folk, 90s Nostalgic Indie" I'll try some other excludes like "rapping" but excluding "rap" doesn't seem to help.

[sleek stump]

Ya none of those should induce any rap in the track. What is your weirdness setting? Try turning that down to zero and styles influence to 75+.

Actually, 90s Nostalgic Indie could get some rapping, theres plenty of popular tracks that included rapping in that genre. Linkn Park would be an example.

[wet idol]

Hey, does anyone know how (or if it’s even possible) to add lyrics to a part of the song that’s just instrumental?

[sleek stump]

Use teh "Replace Section" in the editor.

[little vessel]

I've never really touched the style or weirdness influence setting. I'll keep trying different things but it's a waste of time and credits, I just did one that was simply "Indie Acoustic Rock, Comedy Rock, Narrative, Quirky" and it rapped the whole thing very quickly and took about 4 minutes worth of lyrics and ended up with a 2 and a half minute long song, it seems to want to just generate the quickest it can. Frustrating cause 4.0 and 3.5 didn't do this at all.

[sleek stump]

This is super odd, would you mind linking me the track you are talking about so I can do some testing?

[little vessel]

https://suno.com/s/tM8ugnYV5gMxFN9p

[half dragon]

That doesn't sound like rap to me.

[little vessel]

The issue is the lyrics are being sung quickly in a rapping fashion, not the music style being rap.

[half dragon]

There are a lot of words in the lyric and the style seems to be fast pace so yeah... it sings through it fast.
What style are you going for if not this? slower pace indie rock?

[little vessel]

The style and music is perfectly fine, what I wanted, I just want it to not quickly sing the lyrics. Sometimes it gets it right but it probably 60% decides to do this without being prompted to do so. Whereas in version 4.0 it would almost never do this, it would sing the lyrics proper and not rap them unless it was prompted for rap or hip-hop type of prompt. Very annoying.

[half dragon]

You mean, this same lyric would generate slower singing with v4.0??
There's a lot of words per sentence so with this beat, not sure if it can sing differently.

[little vessel]

Here, I just did the same prompt in 4.0. Which seems to match the prompt better with more proper singing and not just quickly saying all the lyrics. But quality isn't so great. https://suno.com/s/DwuyUW8aLx2O6cSf

Here's another example where I used a more complex style, and everything is going along great, just how I want it, but then at the 2:37 mark or so it just starts rapping a couple of the verses really fast and then goes back to normal singing. The new 4.5 seems to think that's how all songs should happen I guess. Even specifying no rapping in the style, it seems to really want to throw in a rap verse or two in most songs.
https://suno.com/s/bCnTDDXhaqAc6XU5

I've generated probably a thousand or so song attempts in the last 9 months and it only started doing this with 4.5. I guess since they've allowed longer generations it really thinks it needs to quickly speed through some verses, quite frustrating as being able to generate longer songs is a great improved feature, it just seems there are still some kinks to work out.

[sleek stump]

Part of the problem I see is that the styles do not designate what type of vocalist the track needs to have. The entire styles section is super simple, as if you were prompting in 3.5 or v4. 4.5 benefits heavily from very detailed and verbose style prompts. IM going to run this with some enhanced style prompting and see what happens. While what I hear isnt rapping, I can understand what you mean because of the "comedic" and "Narrative" styles you added, thats what is making some of the verses be almost spoken.

Anthoer issue I see is that your verses are waaaay too long. They need to be shorter so the render doersnmt try to jam it all into what it feels is a normal phrase.

[little vessel]

yeah I'll try to use shorter verses from now on

[sleek stump]

Additionally, the genre that you are using are all medium fast, and trying to get Suno to slow those down is going to be a battle.

[hasty pine]

​I remixed a snippet of an original song and I really like how it turned out. I want to extend the remix with my full-length original song, but I don't want to change the remix (i.e. keep the same style, tempo, vocal, etc). How do I do this?

[sleek stump]

Ive run a lot of changes to this, and every version the lyrics are too long and uneven. Suno is trying to fit all of them into a normal set of bars and just cramming it all in , making it fast.

Ive even broke it up into more verses, but they are not uniform so its weird.

[sleek stump]

AT any rate, this styles prompt gets you the correct vocals and arrangement style:

A slow, introspective folk indie groove built on warm, gently picked acoustic guitar, with soft percussion and a playful, bouncy bass line. Light keys highlight oddball melodic turns. Quirky alto male vocals lead expressively, complemented by female harmonies, brief comments, and bright whistle accents.

AT this point all I can recommend is doing a rewrite with verses that are 4 or 6 lines, keep them consistent, make the long chorus into a pre chorus and chorus setup. Add some instrumental breaks and interludes to let the instrumentation breathe. Currently, Its all just long and crammed together so its not going to turn out how you hope.

[half dragon]

Extend would keep most of what you mentioned with new lyrical extensions.; so just simply Extend it and add your lyric for the new part.

[lyric sonnet]

wrote lyrics: "doo doo, doo doo doo; doo doo, doo doo doo doo" - "Can't generate this, lyrics contain copyrighted material."

what the?

[half dragon]

Maybe? https://www.youtube.com/watch?v=D2-uzhyoEyA

[lyric sonnet]

maybe. but interpreting doo doo doo as copyrighted because of that goes too far imho

the lyrics don't even contain doo, it's just the song title

[polar spade]

" ....'do do do, d' da da da' is all i want to say to you..." The Police
" do do do do do, da da da dee da...." Frank Sinatra
Its everywhere....😄

Seriously, though, 'copyrighted" is just a generic term, rather than specific list what banned list of a few dozen it belongs in.

" Content denied due to Common phrasing found which is clearly not exclusive nor creative, so our system doesnt want to train on it ....again....anymore.... and it cuts down on server load" ...... they just shorten it to "copyrighted"

Its tough trying to write children's "under 8" songs that doesnt have the numbers " 1, 2, 3" so I feel for you. Lots of things I cant use.

[polar spade]

Posting here, transferred from Bug channel: "......Haven't been on Suno in a while. Is downloading lyric videos now no longer possible on free accounts? Or some kind of bug?......"

@gray elm
Sorry to nudge in again, but to save time and inform others as well:

Call it a "new behavior" if you wish. The announcement in July of '24 of its introduction to Pro and Premium members didn't initially include Basic. #🚀┃updates message And since then, I've not found that it was ever changed from that. It's only available during the Trial 4.0 (100 credits) period given to Basic members.

As I mentioned, you could be right, by glitch or otherwise, that it was accessible at some point to Basic users. Seems to be a moot point at this time though. The present position is clear at least from this past May (2025) when the following was posted, it seems that video download is not presently an included Basic feature.

BASIC list of Includes and Not-includes: #1373496852112212121 message

Hope this helps from waiting on a response in Bugs 🐛.

[dapper void]

is there a way to control the remastering somehow? i'm trying to update my 4.0 songs, but the results are very random, and i just wanted it to stick to the originals as best as it could

[raven turtle]

Have you tried Remastering your old v4 songs in v4 again?

[dapper void]

i tried, but they also differ from each other too much, so don't get the idea

[polar spade]

Remaster should keep the melody and time duration the same. Not including sound quality and hallucinations, mispronunciations and glitch noises.......Since last year, it does change up a chord now and then, occasionally sings notes flat or sharp unfortunately, and throws in a nuanced riff or instrument here and there, so those are expected. For the most part, though, a 4.o remaster of a 4.o output shouldnt change up too much. But things could have changed.

[polar spade]

Covers are good as well for sticking to the original if you adjust sliders for just that. So a song under 4 minutes, even a Cover might "master" the song more accurately? Does for me if I use it to clean up some artifacts.

[dapper void]

since fiddling with the original music is expected, i think there should be a slider to determine how strong that fiddling would be, because now i need to pray to RNG gods to not mess many nuanced parts of the song. the cover with the sliders usage changes the song even more i think

also i wanted 4.5 for sound quality, as i feel 4.0 is worse

[hearty laurel]

Can someone help me? I am having issues with Suno replacing 1 word in the lyrics of a song. I go to edit, select the word, click replace lyrics, highlight the word and replace it in the box, then hit replace. and it gives me 2 options. when i play either option the word is still the same, it does not change. I have wasted a bunch of credits, and still no result. I have watched a bunch of youtube videos, and they all show how easy it is and it works for them, but not for me. Any ideas? thanx

[polar spade]

1. Change up the content blenders at bottom of the New Lyrics box to try all three, but Smart and Fixed do better for just a word.

2. keep in mind AI is also trained to speak/sing "naturally". So though you may wish to change the word "eat" in "Did you eat", Suno already maybe pronounces the three words as "jeet", so just selecting one word is more difficult for it to make a change. I usually always select the whole verse, or a couple of lines which have enough room on the ends to make any new word fit. As for that, I think the minimum is 3 seconds (not 3 words) anyway, so how do we determine which words fit in 3 seconds, right? Highlighting is too vague.

But if you're using the same words, just needing it to try again to pronounce one word correctly, just highlight the whole verse (if only 4-5 lines), click New Lyrics to make sure the same verse shows up, select Smart , and click Regenerate and see what that does.

[sleek stump]

Using the Legacy editor might also help.

[hearty laurel]

Guys, thanx for the replies. Believe it or not, all I am trying to replace is a typo in "and give this love of our time to grow." to "and give this love of ours time to grow." And Ginger, how do I get to the legacy editor? Thanx again.

[solar sequoia]

I have a another question.
I have been desperately trying to get Suno to make me a plain blues bass and drums backing track I can practice my leads on guitar to. Yes I'm aware YT has such tracks I just wanted to make a couple with this. And no matter what Suno jams a crapload of leads. No matter which genre. Any tips would be appreciated..thanks

[polar spade]

"This love of ours time to grow" takes longer to sing than " this love of our time to grow" so , for Suno , it will be better to highlight more than just a few words or it will tend to mumble.

[polar spade]

LEGACY editor link found upper left of the Edit UI page.

[humble verge]

Persona q: Would the best way to create a persona be to get a vocal stem, crop it (to avoid the composition shining through too much) and create the persona from that? Also, I've been trying to make SUNO give me double-tracking on the vocals, but so far no luck. I wonder if it understands the term. Anyone got any luck with that?

[solar sequoia]

Oh and yet another question I just have this on my android so..
I've been putting together a blues album for my private collection... is there a way I can get Suno to give me an extended jam? I've attempted several like "instrumental extended jam" and "instrumental ad-lib extended jam" doesn't work for my jazz stuff either. The only time I seem to get longer tracks is when I do something in a metal vein, or if I use dubstep in the style box, but other wise I get 3-4 minutes sometimes 5... Not that I want every song longer.. you have to serve the song. And I would love to do some jam band stuff in a Phish or Widespread Panic style.. and a couple of my tunes literally cry out for this. But any help or tips would be greatly appreciated.

[solar sequoia]

Never mind "jam band " works in the style description.. I stuck that tune in the Jazz section if any one is interested...

[hearty laurel]

@polar spade Thanx for the reply. I will try and highlight the entire sentence and see if that works. I remember trying to replace the whole verse, but I was told that 28 seconds was the max. and thanx for the heads up to the legacy editor. I will report back with the results. Thanx again to all for your suggestions.

[stuck obsidian]

May I ask you (or anyone who sees this 🙂) how you get your tracks to post in a playable format like that in Discord?

I use the Android app, and if I simply share the link, it doesn't post to Discord that way.

Also, someone told me to download the song and upload the file, but that didn't work out correctly either. It basically just produced a little play button as a post in Discord. Not entirely sure what I'm doing wrong.

[peak veldt]

@knotty stag I like what you're proposing...

I don't own a DAW, and I am not familiar with them, but if I bothered to go that route to download a DAW, and figure out how it works, it be a worthy suggestion you have there.

But I rather have SUNO generate what was asked of it (within a few tries) so it can stay in the Suno platform playback system. Once I go DAW the only way to get the genie back in the bottle is to upload back into Suno. And that's a bit too much to do to get duets to work properly at this time.

[little vessel]

Thanks, that is helpful, I mostly use chat-gpt for lyric generation so I'll make sure to try and get it to prompt things like this better.

[brazen yew]

When it comes to prompts, do people find more success with just a list of desirable aspects (genre, vocals, description) or making it structured. So you list everything separately.

[brazen yew]

Maybe try copying the link from suno in the mobile browser?

[sleek stump]

That really depends on teh model you are using, 4.5 wants very detailed style prompts, previous models is better to keep in more simple.

[brazen yew]

Ok thank you, I'm using 4.5 so I'll try and be more detailed. Have you had any experience mixing the description into the lyrics aswell to almost prompt suno twice to try to get more accurate hits?

[sleek stump]

There is a lot of nuance you need to learn, and making use of the magic wand (upper right corner of styles box) can help yo ua lot withj 4.5. You will also want to reinforce what you are describing in styles with meta tags in the lyrics.

[stuck obsidian]

I'll give that a try. Thank you!

[brazen yew]

Ok thanks. I don't have a magic wand on either the suno mobile site or the app unfortunately

[sleek stump]

Ah, yes, mobile apps are limited in features.

I would highly recommend using a desktop or a laptop and logging in to the website, you can learn a lot from using features like the magic wand.

[main gulch]

is the order mixed up in the trending lists?

[brazen yew]

One frustrating thing is it seems if I put piano as an instrument amongst others the whole song becomes 80% piano

[old hinge]

Has paying for subscription working for Android yet? Ik last I tried through cash app and it didn't work but that was during when it was only a day or two old of released on Android

[raven turtle]

Considering the 50 credit stem split option for a moment , has anyone done and comparisons on whether the summed together 50 credit output, 10 credit vox/music bed, or flat out WAV are the same fidelity, or do you get better quality out on a stem split?

[graceful condor]

Anybody know what exactly the asterisk is supposed to do in suno? I used to pronounce the word more? You are *marked* — in stillness. But it skipped over the word marked.

[hardy thunder]

I have used asterisks to create sound effects like *gunshots*

[tough isle]

I haven’t figured out the best way to use Suno to create a song that fully—or at least mostly—sounds like my own voice. I use the upload feature to upload my voice, and even though I turn up the Audio Influence setting to the maximum, the result still sounds very different from my real voice - any idea?

[graceful condor]

I do noticed in the editors lyric window it's only show one word where I used the asterisk but I have 7 words where its used and it skips right over most of them.

[gloomy shore]

Anyone knows if Suno has audio visualizer feature?

[polar spade]

Some use them WITHIN parentheses for FXs.... but alone, I haven't seen any consistent verifiable purpose.

[gloomy shore]

Is there an option that you can download the audio as a video format that has visualizer?

It's alright, thanks

[neat nexus]

How did you prompt it to avoid areggio? If you put that in the style prompt or lyrics prompt you basically inceptioned it to do more arepeggios.

It only works if you put it in the exclude styles section. And even then it can be a bit hit or miss. Remember it's AI. It's going to try to think on its own and that means making decisions we may not be happy with.

@wise kindle

[polar spade]

@graceful condor You mentioned in need-help channel 4 days ago that you couldnt download WAV files. This seems to be a universal glitch. That means there would likely be no individual response from Support in that they would simply focus on the solution for everyone. Its a bug to be resolved. No individual "support" needed.

But you said its "been a month" of an issue. I didn't find another post from you but the one 4 days ago. Have you emailed support a month ago about another issue?

[keen sphinx]

hi everyone, not sure if it the rigth canal for asking that. but i would need advice/tuto on how to do thing. i really like one of my suno generated song, but i need to correct some lyrics and the voice of the singer. but i dont know how to make sure suno gonna use the same instrumental base and change only what i want to be change

[half dragon]

Its really simple - load the song into the Editor > Replacement Section
Then highlight the word(s) in lyric and type in the same on the new lyric box , then hit Replace
Listen to the generated replacements and decide
Rinse and repeat.

[keen sphinx]

thank you i tried some thing in the editing mode, but when it come to the finishind part i end up with a product who not smooth in the way you can hear clear cut between the old and new generated part. how do you harmonize the whole ?

[sleek stump]

The answer is no, you will need to take your track outside Suno and make a visualizer video. I currently use 2 different things for making music visualizer videos, Renderforest being one of them.

[keen sphinx]

also i'm trying different setting for the lyric remplacement, like smart/classic/and fixed. but i dont know why fixed mode give me yogurt english and not my lyric at all XD such a waste of credit

[half dragon]

I mostly keep it on Smart - Have you tried it?
Even on smart - sometimes it generated gyberish as well - I just tried again

[keen sphinx]

yeahh i got back on the smart setting and it finaly give me somethign good

oooook , now i ask it to add a part with new lyrics and he just ignore it totaly and create a part with lyrics from the chorus, it gonna make me crazy

[keen sphinx]

how would you do it to specifie to suno the rythm and "emotion" to put on specific part of your song ? like be soft and verse 1 angry on chorus etc... ?

[zinc cloud]

Haven’t used the app in awhile does it still create lower quality songs compared to the website? I find that the website produces better quality songs

[radiant ridge]

Anyone got any tips for making Suno NOT creating a guitar solo/lead? I am trying to create myself some blues backing track instrumental. But for some reason Suno always want to add guitar solos over the entire track :/

[raven turtle]

Suno is having one of those days where it only serves me reject Eurovision entries. I'll give it a rest for now as the weirdness slider is off the charts! LOL

[polar spade]

@jade geode ...... you ran through 2500 credits in about 5 days? Been there. Done that.
You might consider.....For $30, though, you can get 10,000 credits and upgrade to Premium immediately, even if for just 30 days. Just remember to downgrade within 20 days (before 7 days to renewal) to set it to renew at $10 Pro rate again, or else get charged $30 Premium again at renewal date. At renewal, reminding you, all unused credits will refresh (topped off) then at 2500 for Pro again.

Oh, and if you do, dont downgrade immediately. Give a few days for all funds to post, for your receipt to come in the email, and you're smoothly upgraded and rolling. Set an alarm, a calendar appointment, a sticky note, whatever it takes, to remind you to cancel no less than a week before renewal date.

[polar spade]

Thats called a baited question, my friend.

Lots of folks having a blast using Suno. So, its either a yes or a no, right, depending on tastes. Are there still issues. Yes.

The best response would be the average quality continues to be going in an upward direction.

Just read through the posts and listen to trending for current stuff. You be the judge. Good luck.

Why are the random test ones always the 8 minute songs. 😅

[keen iris]

I need some help with Pro Plan, I have near 270 crefits, but Suno says that I don't have access to 4 and 4.5 versions.

My subscription ends today, but today is not end yet, right?

I'm in eastern Europe so here is already 14, but from my expirience with Suno I know that I have few hours.

Correct me if I'm wrong and also advise me if my Pro Plan ends on my local time, than tracks I made with 4 and 4.5 versions after are mine or not?
Sry for my eng

[polar spade]

The subscription typically is "triggered" to renew on the date and time you had originally purchased. From personal experience, my renewal times have varied within about a 6 hour range.

If it says renews on the 14th, though, that should be your date. But to check that, just look back (or check Manage account) and see what date you purchased.

You always own and control the outputs you make in Paid accounts, even if your account is expired or closed and you leave Suno.

But first, it would RENEW today and not END. Unless you clicked CANCEL and confirmed a week or so ago that you are canceling to drop to Basic. If you did, It would also show a "Plan End Date" in your subscription profile.

[keen iris]

It is canceled, so no renew.
But seems it gives me create few tracks after, because I have last tracks on 00:19.
However it shows that my current plan is Basic but credits are still there, is it right?

I mean is it how it should be?

[polar spade]

There is not really an "Expire" with Suno. Often many users will let it renew with no funds in a bank account and assume that will stop or "Expire" the account. Stripe, the payment processor will continue to ping the bank/charge account a few more times during the week , and can lock the account.

[rugged stirrup]

Question if you pay for a subcription and that subscription expires after a month and you didnt use all the credits it offered

Do you still have those credits or do they go away until you renew the subscription

[polar spade]

Meh. Shrug. 🤷 who knows. If it renews to Basic, it would likely automatically give you 50 credits at least. But , you would not have access to Paid tools of 4 and 4.5 versions. There may be a delay, as you say, so enjoy while you can.

[polar spade]

Subscription automatically renew and your funded card gets charged again on that date.

The credits you get are topped off or "refresh" back to the maximum 2500 for Pro or 10,000 for Premium. There is no carry-over or roll-over. Remaining credits do not accumulate month to month.

[keen iris]

Thanx, but I like to enjoy it when I kinda own it😅

[polar spade]

I have seen some accounts "drag out" for about 3 days on a Pro account past the renewal date, while a payment is floating to be posted when someone changes their payment card, according to users stories. True or not. Who knows. So, things seem to vary, now and then.

[polar spade]

If any time you bought extra credits, those go with your account and can be used for any subscription level by the way . They never disappear (since about February 2025) while your account stays open. [edit: UNKNOWN at this time]

[keen iris]

So If I buy extra credits, I can use them after Pro plan ends?
As long as I know customer have rights only on tracks created while subscription is active, maybe I'm wrong

[sleek stump]

I dont know about current, but any extra credits you purchased while subscribed (earlier this year), could only be used while subscribed. The free account did not allow usage of those credits (those that never expire)

That may very well have changed, but I don't really know if it has.

Additionally, you can only purchase extra credits while under a current subscriber plan, you can not purchase them on a free acount.

[keen iris]

It's a shame, I d like to buy credits and not subscription, because I have not enough time to spend all 2500 credits in month 😭

[polar spade]

Yes. You may be right then. And so id be assuming now to say we can use extra credits even on a Basic account. Dont know at this point if we can or they just zero out. Again, I think they made changes back a few months ago, but as most, they are simply things to be soon discovered to be changed without notice.

So we gets lots of Mandela effects in here. Many still swear we could download videos before while in Basic mode. (Now we just sit back and wait for comments. . . .) 😁

[polar spade]

But just takes one volunteer to find out. Hahaha. [I've tired of losing my credits being the guinea pig myself]

[icy spruce]

/create

[pseudo blade]

use suno app/web

[sleek stump]

careful, Woodhouse might create something you wouldnt want to touch 😛

[polar spade]

@pine yarrow From your Need-help question:

"....
Is it possible to transfer my subscription to another account? Somehow instead of renewing my subscription, payment was taken and started a whole new account. Which is not ideal since I have a ton in my library in my original account....."

According to the Suno Knowledge Base https://help.suno.com/en/articles/2553409

. . . . In any case, if you do not see your previous songs in your Library, there's a good chance you have more than one account. Contact our team at [e-mail] so we can help you navigate and recover your account details. We can also help you merge accounts if you have songs in two or more accounts!

[random stag]

Is there a way to generate sheet music to print out so musicians can perform the songs live?

[pine yarrow]

Ty but I reached out over a week ago n still haven’t heard back from them telling me they were gonna look into it.

[half dragon]

Not on Suno; but you could try 3rd party service such as: https://klang.io/

[polar spade]

I assume for non-urgent (comparatively) issues, response time can be as much as 5 weeks from Support. Just a guess.

But as for your question if its possible, it appears it can be done.

[vital cedar]

mv

jh io p\

ioh

oh ]oh

pij

ih

[graceful condor]

Then why couldn't they just reply like that when I contacted customer support? Is it their job to decide on their own whether individual support is needed and simply ignore the inquiry without saying anything? I've never heard of a company that handles support like that. Also, the issue I contacted them about a month ago was a completely different matter. I never posted about it here—I contacted them via email.

[polar spade]

I have no idea how Support works with Suno. I hope you get whatever answer you look for. Im just giving my opinion, so maybe not worth much, right.

Often most companies , in my history of things, take (triage) all the common issues in emails and put out a single update now and then for everyone. This avoids delays in answering all similar emails with the same message: "We're on it."

Nonetheless, hopefully, as downloading WAV files is a pretty universal need, it is a higher priority than most and is resolved soon. Fingers crossed. The workarounds are helping many, but not everyone, as you say they dont work for you.

Seems like that completely different matter of yours would be the main priority. Check back in the need-help channel in about 7 hours to see if a moderator is hanging out and bring your issues up again then. Being patient but persistent is always a good thing.

[frigid orbit]

Thought I’d drop it here since team/mods seem to frequent this more

[upper heron]

i’m trying to keep this as close as possible while making the make mix clearer and cleaner , i’m trying to keep distortion on the bass only mostly

https://suno.com/s/P965N3PwwxrWmDBe

i’ve tried a lot of things so i’m trying to get a fresh take

then i’m trying to put about 3000 character worth of lyrics (it’s one long verse) over that and to be the overall mix to sound clean as possible

[pseudo blade]

make sure to have a song structure tag so that means only 2993 characters only

[agile orchid]

Hey all, wondering if someone can help me. I have a song idea where I'm singing a chorus. Is there a way where I can input the song to Suno and have it keep the exact melody but re-sing it better (even with different voice) ? I've tried extend feature to do this, and can't seem to stick to the exact melody

[half dragon]

Try Cover instead.

[agile orchid]

thanks! Do covers maintain the BPM and key of the song?

[half dragon]

Cover maintains melody - if you choose a style that matches the uploaded audio, it might be able to have the same bpm or close

[frigid orbit]

You can change BPM in the editor. Depending how far your cover prompt differs from the audio, it'll either be close or far.

[sturdy dove]

@daring lava sends me here with my question...

Alright, so I'm trying to edit my song and the Editor is clearly still in Beta because it's incredibly difficult to work with. The main reason I'm having issues is it's completely misinterpreted and lost major sections of lyrics.

I see it's possible to realign in the Legacy editor, but I haven't figured out how to do that yet. It's not particularly intuitive. Is there a way to re-align the Editor (and sections) with lyrics? I can just keep trying and trying if I need to...

I spent my entire day off yesterday trying to make some minor edits and I'm very close to the result I want with just a couple of minor tweaks. I can't make those tweaks because the lyrics don't exist in the editor and the sections are all misaligned.

I'm extremely frustrated. Anybody have any ideas?... THANK YOU!

[hard beacon]

@polar spade : i know that you know a lot 😄 😄 do you know if it's possible to change only specific parts in a track?? imagine you have a great dubstep output but you wanna re-create the drop... is that possible?? don't work that much with the editor.. thanks 🙂

like a "new drop" function 😉

[polar spade]

@sturdy dove The simplest answer is to click "align" or hmmm.... "Match Lyrics to Audio", actually. Watch the purple highlighter as it indicates words matching to audio. It should be word for word.

If several audio words are sung but the highlighter just hangs on a word or space, we compensate in the Corrected lyrics side by adding those words or spaces or anything needed to eventually get the highlighter to match word for word. I have had to save changes, look at the results again , and repeat the process until everything seems aligned -- before make any lyric changes.

It IS a learning curve and seems to take finesse, but I'll add, Suno does occassionally and recently re-arrange the final lyrics. This makes it frustrating, if we're not looking for it, and just means a couple more alignments. [ Its like using DOS instead of Windows -- Suno may have a GUI appearance, but we still have to understand its gremlins to keep ahead of it]

I have a habit of doing this before any lyric change.

[polar spade]

Not on a consistent basis. I, myself, would try duplicating the drop, splitting the exact spot in the output i want it and squeezing the duplicate into that split. Then Cover or remaster and see if it sounds good or just leave it alone until youve made all edits and remaster later.

Not something I've specifically done. I've only duplicated whole choruses. So doing the same , but more surgically, would be interesting if it worked.

I've personally added drum beats , or extended measures using duplicates, or manually splice in or out single wave forms to emulate a beat or sound, similar as one would using a DAW. We just zoom in as far possible to do those and eyeball as well and constantly check audio.

A drop would go for a good half-second though which is quite long, so harder to see the exact spots to place (harder to visually zoom in) , as well as we'd hear more of a click at the end. Remaster with AI though can do some fun blending.

Im about 1% smart on editing, so hopefully others can shed some light on their ideas, within Suno, (before just giving up and going to DAW) on how to do it.

[brazen yew]

Any metal creators who have managed to keep their song clear of clean vocals?

[sturdy dove]

Yes, that makes sense! (excuse me, let me clarify:) *But it is not work like that practically. I've actually just spent a few minutes (half hour?) adjusting all of the sections into broken and not broken, color coding them, and titling them properly. Unfortunately, no matter how many times I "reload lyrics" it doesn't fix anything. All of the lyrics are completely out of whack.

Without the ability for Suno to detect the correct lyrics in each section, no useful edits are possible. I blew 100 credits last night trying to replace the broken sections with new lyrics, and the first few came out as intrumentals, the next bunch came out as garbled gibberish.

I was hoping that with all of the sections broken out properly, it'd be able to reload the lyrics one section at a time and figure them out. But so far, no dice.

Thanks for your help by the way!

It may be that what I'm trying to do is simply impossible. It's ultra-frustrating because I'm 80% to where I want to be. If I could get another 10-15% it'd be "good enough" and I could then continue to work on it later, as Suno improves its tools.

[polar spade]

Certainly go over and try the Legacy version. For some it just seems more intuitive or just works for them. Who knows. Sorry then that its not working out. We need some zoom classes or something in all this, eh?

[sturdy dove]

Okay. Bummer. Okay. I'll save a new version / song of what I've done and pull it into Legacy.

Bah. SO CLOSE.

Is this something an external DAW could help with? I used Kits in the past to remaster an Udio song, and I've seen some people suggesting it, but I'm not sure if it's right for this task.

If I could edit individual stems, I think it would be way easier for Suno to figure out what is needed. I could just keep extending lyrics in that stem alone until I get what I want, then plunk that in somewhere. (I did that with an Udio gen long, long ago, I was able to take the brass section from what I was working on, regenerate it, and paste it in using some basic tool, then recompose the whole thing and it worked brilliantly.)

[sturdy dove]

BAH. None of the edits SAVED in the new save. Holy mackeral this is frustrating.

[polar spade]

DAW work is a totally different thing. Sure. So using DAW and applying the use of a dozen other AI sites, im sure editing would be more a breeze.

Im just focused on what can be done here in Suno to help everyone in the spectrum from new to experienced.

So, I dont have too much issue and get an exceptable output when I edit. So it could even be that I have less instruments and vocals within each selection. My genres are folk and pop, often. Metal , rock, heavily orchestrated pieces, epic-core songs, all those may be harder to align? Who knows.

[sturdy dove]

Yeah, I'd greatly prefer using Suno standalone, but it seems like what I want isn't possible. I've invested a huge amount of time and am no closer to an acceptable outcome. It's frustrating. Perhaps the internal tools will improve in the future.

[mental topaz]

Question, I am noticing with my prompt, that it's coming out in the songs for example "let the chorus swell" - nowhere in the lyrics but I use it in my prompt, they will add that to the song and sing it. How to stop this?

[soft arch]

I've only had that happen once, so I considered it a fluke. Be sure you're putting your [directions] in square brackets, and be sure you didn't leave a previous direction open (missing the end ] of previous directions). You might try simplifying the direction, too, *[chorus swell] * instead of [let the chorus swell]. I usually have better luck that way. Beyond that, I have nothing.

[sleek stump]

You want to keep meta tags as simple as possible. Overly long or complex tags usually get ignored or even sung because Suno doesn’t know what to do with them.

[polar spade]

Is that in your style prompts or in the lyrics box ?

[hearty laurel]

Ok, guys. I need some serious help from someone at Suno. I have WASTED over 300 credits trying to change one f***ing word in a verse!!
I tried replacing a section, tried the entire sentence, tried the entire verse and It still will not change the word or lyrics! I want my credits back!

[soft arch]

Need more details than that.

[hearty laurel]

what do you need?

The line that was typed in the lyric box was "We don't want to rush things, let's take it real slow,
and give this love of our time to grow." I just want to change it to "this love of ours". How difficult is that?

[soft arch]

What word are you trying to change? What are you getting for results? What have you tried to do? Are you trying to change just one standalone word, or one word in a line or two? Stuff like that.

[hearty laurel]

DennisG, thank you for the reply. With over 300 credits, I tried everything! Tried replacing the word, still the same. tried replacing the entire sentence, Same thing. tried replacing the entire vers, same thing. it still sings it as our and not ours!

Even the lyric box shows the new lyric, but it still gets sung as our.

[soft arch]

Open the song in the editor and hit the reset button. Go back to the create page, then into the song details and make sure the lyrics are correct. Sometimes the words are messed up there and Suno will pick up on that. Turn off all filters and delete all the previous attempts at changing the word. Reload the Create page. Open the editor again. Highlight the line with the word you want to change. Open the the panel to change the lyrics. The line should be populated in that box. Change the one word and give it a shot. That's worked for me when Suno seemed "stuck" on one way of doing things.

[hearty laurel]

DennisG thanx for the info. Unfortunately, I like the audio and the singer as it is now, and I don't want that to change. I just want the word replaced, that's all. I will try your recommendations and let you know what happens.

[soft arch]

If you do like I suggested, I don't think you'll see much change other than the one word. That's what has worked for me.

[hearty laurel]

Ok, I have tried all you have said and now it is singing the entire line differently, not the way it was before the change! The word sounds like ti could be different, but the entire line is not correct.

[soft arch]

I don't know what else to tell you. I did that earlier today and it worked for me. I might have regenerated it a couple of times, not sure because I did other changes, too. I haven't tried this next idea, but you could try changing the "auto" setting and choose the setting for "something short" (don't remember the exact wording, but you'll know it when you see it).

Or you could follow my previous instructions and just highlight the one word instead of the whole line and see if it will get it right.

[hearty laurel]

No luck. I have gone from 3100 credits to 2600, trying to fix 1 stupid word!!!!!!!!!!!

[soft arch]

If no one else here has an answer, try contacting support.

[elder dew]

Are there any prompts or best practices to help with the muted/underwater sound that 4.5 puts out when doing metal song gens?

[hearty laurel]

Ok, sent an email to support. lets see if they are any help.

[boreal harness]

Split it. Get the Vocal stem. Edit the "displayed" lyric if it's wrong. Remaster just the vocal stem. Usually works for me.

[sturdy dove]

I'll chime in here: maybe try emphasizing that last 's' a little more, like replace it with two zs. I tried to get it to say the letter 'h' and it just kept ignoring it until I spelled out 'aitch'.

Replacing the entire section is (yes yet more costly) but also likely to yield results you'll like. But yes, the editor definitely needs work.

[azure crane]

Can't generate Stems,It's stuck in "Detected Stems" mode.

[polar spade]

Is it still stuck. What happens if you click "version 2"? For some reason, my version 2 loads faster than version 1.

[azure crane]

It stucks with “Detected Stems” mode but the stems are generated cause I can see them in the archive. The thing is that I have to d/l one by one which is a big waste of time.

Hmm, my subscription is expiring soon so I was planning to create a bunch of stems… but now that I can't create the two zip files I won't be able to use up my credits…

[azure crane]

This is really crucial. I have created a bugs msg but they really have to fix this now…

[safe snow]

Hello! Any tips to keep voice consistence? Like having a duet singing different parts of the song and choosing what voice sing each part

[sturdy dove]

Uh, good luck with that. I'm in the middle of dealing with this. Expect hundreds of generations to get close to what you want. And the editor won't fix it. Just have to grind and refine.

[sturdy dove]

Of course clear prompts help, but that's only the beginning.

[safe snow]

Did you found any method that seems to help? Specific words to prompt or metatags on lyrics for example

[sturdy dove]

It seems to interpret pretty much anything as well as anything else. One tip that I think is actually useful is a longer tag before everything that explains what's coming is a duet and what kind of duet it is. Let me see if I can find the prompt that I used for my more complicated duet...

[sturdy dove]

I can't get at it right now, but the one that seemed to perhaps help was a bit complicated, I found it online somewhere. Something like:
[ Duet between a male vocalist - Vocals 1 and a female - Vocals 2]

Then,
[Verse 1 - Vocals 1]

or
[Chorus - together in harmony]

or
[Bridge - Alternating]

Again, none of this works all of the time, but I seemed to get more consistent generations with these tags. Again, I did at least 100 to get something remotely close to acceptable.

[safe snow]

Thank you very much for your help

[random stag]

Do you think Suno would entertain a partnership with one of these?

[half dragon]

I doubt it. But who knows.

[buoyant gale]

I have tried this 3rd party app in transcribing audio to sheet music and they are not accurate. Klangio is not free, but there are many apps that can do this for Free but they're not accurate you still have to manually edit the sheet music, but at least it is free though

[short sigil]

Hey everyone, what if my distributor asking for a license?

[polar spade]

What did they say was a license? If they cant show you, how would you know?

Does this distributor happen to be Too Lost? They've been asking for licenses lately it seems.

[short sigil]

They just blocked one of my tracks and asked me for a license for mastertrack or something

[polar spade]

Right. They leave it up to you to give them something they are pleased with. Problem is, there is no official USA Music Industry boiler plate transfer of song ownership commercial license.

So where one distributor accepts what you give them, another rejects it.

[polar spade]

There have been a few if not dozens if not hundreds of AI app users who have submitted songs and had them distributed.

Presently you have a "commercial license w/terms and conditions" written and agreed by Suno and you, found within the Terms of Service. It is a general contract that covers any and all songs created (within terms and while in accordance with US laws) and to be created.

[short sigil]

Yea, i finally send all that i have to them: suno terms, my receipts

[polar spade]

Is this "Too Lost" distributor? They seem to be asking recently.

[short sigil]

No, "TuneCore"

[polar spade]

Ahhh. That makes sense as well.

[polar spade]

Ill wager as this goes on, many may start asking for a specific license, than a general one. A specific license would have both parties names, and included the name of the song as well as any other details regarding its creation.

SUNO is never obligated to provide a specific license as that, in the realm of AI generative outputs would be ludicrous to list n-number of songs per person and more created every day.

[short sigil]

That's make sense

[polar spade]

The bottom line is that if its unclear to Tunecore , they dont wish to be liable if someone comes after them for any infringements found from your song. They want a license/contract that clearly holds you 100% liable.

[polar spade]

Oh. I was curious. Did some tracks go through , but they blocked others?

[short sigil]

Yea

[polar spade]

If both the ones that went through and the one that got blocked were all AI assisted, what was the difference in what you sent to get some through and this one not go through?

Just .... time?

[copper marsh]

Hi Suno team!
I created a track in my account, but I can’t locate it anymore under my creations or history.

[polar spade]

Both library and workspaces have filters now. So you may need to click off some filters to show more files.

At minimum , wait 20 seconds and click refresh. The server could be overloading and lagging.

[hearty laurel]

@sturdy doveThank you for the reply. I tried your suggestion and replaced the "our" with "ourss" and that worked, but it changed the way the line was sung. So I created a new vocal stem, downloaded it, and then added the new vocal stem to a new track, deleted everything except for that word. Then I blended the ending of the word with the original that didn't have the 's' at the end. Sounds like it was always there. Thanx again.

[azure crane]

I don't understand why the stems separation hasn't been fixed yet after 2-3 days. Has anyone among the technicians noticed this? Since it stops at the “Detected Trunks" procedure without creating the two zip files. Then having to download the stems one by one makes it so time-consuming, like in my case when I want to create many stems so as not to lose all the credits I have left... Very frustrating.

[pseudo blade]

Try using stem seperations externally

[sturdy dove]

Woooooo! That's fantastic! Wait, what did you do?

How did you add the new vocal stem to a new track?

I think I get the blending part...

Was the new track a fresh generation or something you had?

I'd love to know how to add vocals to tracks! I can't get that to work in Suno for the life of me...

[azure crane]

Hmm, Suno’s stem separation is so much better than the external alternatives..

[pseudo blade]

well i'll never know since im a free user but i hope they fix it soon

especially when saturday to sunday isnt a work day

[buoyant gale]

So it shows they can now detect if the songs are AI generated or not.

[hearty laurel]

@sturdy doveSorry about the confusion. I forgot to mention that I added the new vocal stem to a new track in my project with the original song in my DAW (Reaper), and that is where I mixed them together. I apologize for that. That would be a great feature if Suno could do that.

[sturdy dove]

Ah, now it makes sense! I'll check out Reaper...

[lunar nest]

Hi everyone! Is there a way to get Suno to generate songs using the barrel organ or theremin? I've been trying for two days. I managed to create a theremin-ish sound with complex prompts, but most of the time it doesn't work, and explicitly using the word "theremin" in the prompt often somehow ends up producing metal music. As for the barrel organ, I've had almost no success. The sound almost always comes out as 8-bit or like a church organ. I've tried "barrel organ", "street organ", and "orgue de barbarie", but nothing changes. Using "mechanical music" just makes it worse, with 4.5 it ends up sounding Always like 8-bit/electronic music. I've experimented with all available versions of Suno, and in 4.5 I've tried Weirdness at 0 and Style Influence at 100, then Weirdness slightly higher, then both at 50. Am I doing something wrong?

[sleek stump]

You can get a theremin somewhat easliy by describing it in your styles prompt and adding 60s Science Fiction Movie, eerie, and classic space movie as a style. Using the magic wand could further enhance the styles prompt, but I would suggest trying it both unenhanced and enhanced.
Also, theres really no wrong way here, you just need to understand what is possible and know what to ask. The problem with instruments like the Theremin, theres not a lot of use in songs outside sfi-fi movies for Suno to be trained on, so it struggles with that, much like some irish instruments, Suno has no clue what they sound like because of lack of training.

[brisk flint]

Hi. I suspect that Suno doesn't have a ton of training data with a theremin, so it might be helpful to describe the sound very precisely in both the style and lyrics prompts. I just tried it and it worked on the fifth attempt. Feel free to take a look at my prompt: https://suno.com/s/0YDI4vgdXBaNOj4d Have a great evening! 😗

[glass drift]

Has anyone managed to make a song start from the end of another? Two different songs, but they end and starts seamlessly. Like pink Floyd did with the dark side of the moon.

[sturdy dove]

So Dasvon, thanks again for your offers of encouragement and support.

Here, you can see the detailed cut up I did of the song. And you can see, I have a Chorus selected, but the lyrics highlighted are part of a verse. No amount of "Reload lyrics" makes any change. And when I save as a new song, it reverts to the old version without all the careful separations and color coding - imagine instead of 15 sections about 5.

Do you have any ideas for how to make the new sections stick? I've tried saving twice now and it reverts. I could just keep trying to save - maybe it's like generations, sometims it works, sometimes it doesn't? Or maybe there's some trick I'm missing...

A couple of the sections are straight up repeated because I want to try to replace them but I don't want to lose the original in case the replacements don't work. I've tried adding fresh lyrics to the new sections and it doesn't go well. I'm thinking I'll go the DAW route, but I'd rather see if I can figure this out - and see if I can get Suno to see what I'm doing. It seems to be ignoring me...

[scarlet python]

Hello! We have paid for the Pro sub and would like to know how we can upload our songs to other platforms. What do we need to do in order to do this?

The information on this site - https://suno.com/terms - heck, nothing is clear. Or am I stupid? How do I distribute music? How do I confirm to the service I'm going to use that I have a license? Is there any information that can be sent to the distributor?

Do I even need to prove that to the distributor I have a sub to Suno?

Please help😭

[forest flower]

Lets say you extend a song with vocals. How come v3.5 does a consitantly way better job than v4 in terms of (example) mimicing the sound of how it sounds in the uploaded reference song. v4 is way too creative sometimes. Although v3.5 has a more poor vocal quality, it gets the vocal mimic and overall sound of the song nailed way better than v4

Ignore the Rap Verse, and hear how (and I believe), (master quest) or (Pokémon) sounds like in v3.5 compared to v4

[pseudo blade]

You can distrobute into other platforms. If you're pro or premium you should be fine, but for a free user, you cannot distrubute. You can use distrokid or routenote

[scarlet python]

I have a premium, everything is fine🥹 But what should I do if the distributor asks for a license?🤡

[pseudo blade]

well dont be worried..

they wont know

right?

[scarlet python]

shure 😊

[empty steeple]

hello, i’m not able to save my song as a new song from the editor. i click the button and i get the spinning wheel but nothing happens.

Subject: Pro Plan (8,835 Credits) - Save Song Failing in Chrome and Safari (HTTP 500 Errors)

Hi Suno Support,

I’m on the Pro Plan with 8,835 credits and unable to save songs in Chrome (version [e.g., 126]) and Safari (version [e.g., 18]). Clicking "Save Song" shows no confirmation window ("Go to Song" or "Song Saved Successfully"), and songs don’t appear in my Library. I’ve tried:

• Clearing cache, cookies, and local storage in both browsers
• Logging out/in
• Confirming 8,835 credits available
• Checking Library pagination

Console errors include:

• "Failed to load resource: the server responded with a status of 500" on studio-api.prod.suno.com/api/studio/render-state
• CORS error: "Access to fetch at 'https://m-stratovibe.prod.suno.com/agg-receiver-service/v1/events/m' blocked by CORS policy"
• "Error sending segment performance metrics TypeError: Failed to fetch"
• "SyntaxError: Unexpected end of JSON input" (local storage)

Account email: [your email]
Platform: Web (Chrome and Safari)
Device: [e.g., MacBook, iOS version]
Please confirm if this is a known server issue and provide a fix timeline. As a Pro Plan user, I expect priority support.

Thanks,
[Your Name]

could you please help me get this fixed so that i can save songs as new songs from within the song editor?

thank you!

[sleek stump]

Hate to break it, there is no support here. What you posted above should be in an email for support. If this has just happened it could be due to internet issues, server restarts , temporary web services outages, or a number of other temporary reasons. If the issue persists over hours or longer you should likely email [e-mail].

[hearty laurel]

Ok, new problem, guys. Suno is stuck on detected stems. Tried twice and they took 100 credits, but still no stems. How do I get my credits back, or get my stems???

[unique stone]

Either go to your Library and adjust the filter to untick "Hide Stems", or try refreshing the page and going back into Get Stems then "Show Extracted Stems"

[azure crane]

But it still not creating the zip files. Now I have to d/l every single one stem one by one which is so time consuming cause I have a lot of tracks that I want to create stems from… So annoying… They have to fix that bug….

[hearty laurel]

@azure crane Ditto!

And not only that, but out of 11 stems, there were 7 with nothing in them!

[unique stone]

For reference this is what happens when I try to get stems (first screenshot). It gets stuck like this. But if I wait a few minutes and then close this, refresh the page, go back into Get Stems and "See Extracted Stems" it shows the stems and allows me to download them as a ZIP.

[hearty laurel]

That's great! But does it charge you another 50 credits to get the stems? It cost me 100 already with no stems.

[half dragon]

You won't be charged again if you click on "See extracted stems"

[hard beacon]

hmm..

Are the servers hanging right now? My outputs are running and running and running the “load progress”.

ok.. maybe "hanging" is a german to english translation.. maybe the word "lagging" or "down" is better 😄

[neon vortex]

Well it got real slow all of a sudden and wont generate anything for me, so...

[hard beacon]

same here.. ok.. then it's not only on me 🙂

i hope it won't eat my credits then 😄

[neon vortex]

It doesn't. It tells me that no credits ware used. Even if it shows credits as used, if you reload the page it will revert back to whatever credits you had prior to attempting to generate

[hard beacon]

ok.. the general-suno chat is full with "server" probs.. so I could have answered the question myself 🙂

[azure crane]

Wow, I should have thought of that... Yes, you totally saved my day 😍🙏.

[azure crane]

Hmm, well it only seems to work with the latest stem generating… Not the once that I have already generate.. well, better than not working at all…

[polar spade]

I got about 36 stems into it before I figured that one out myself. Haha.

[empty lake]

Typically need to define a specific metal genre in the style description. With 4.5 death metal is normally the best bet, but your mileage may vary. You can try creating a persona with a track that only has growled/harsh vocals and then molding new tracks from there, but, again, your mileage may vary. I suggest putting in a few obvious prompts into the style description and then hitting the wand for Suno to create a more cohesive description. E.g.: death metal, growling vocals, aggressive guitars, etc. If you want to create a non-death metal track with harsh vocals, it will be more difficult.

[brazen yew]

Thanks for the reply, I'll take all of that into account

[waxen wigeon]

i did it just by going to upload audio i just uploaded a song created on Riffusion, because i want to create a persona for it .it would not let me do that, but the upload was not a problem.

[sturdy dove]

Alright, I need to work through this since Suno's editor isn't working in any useful way and I need this project completed.

I'm thinking I'll take the vocal stems, isolate the male and female voices, and feed them to Kits. I'll then have kits make them sing the correct lyrics. I'll patch in the corrected lyrics using Audacity and adjust the timing until it all works together. Then I can have Kits remaster the whole thing...

Does this sound like a good plan? I don't have any other DAWs or tools currently. I'd have to get a Kits subscription.

I'd very much appreciate hearing from the pros! Thank you!

[polar spade]

I recently use extensions+personas now to get a male or female to sing a specific verse. Outside of that, I know I'll have to splice and then blend the splices here and there.

Maybe that's not for everyone, or it may not work for everyone, but seems OK to me so far.

[sturdy dove]

Cool! Thanks for sharing! Do you mind waking me through your specific workflow, or do you have it posted somewhere?

[quaint whale]

Part 1: The Damnation
Part 2: The Summoning
Part 3: The Suffering
Part 4: The Collapse

https://suno.com/s/oEkcPkchagwf2TBj
So this song was created with 4 parts of 4 different songs. However, I was unable to simply bring another song into the Suno editor (as it would only allow me to edit 1 song at a time, so I had to bring them all into a DAW, stitch them up, then bring them back to Suno to smooth some of the stitching process.

1. Is there a way to bring multiple songs into the Suno Editor and I just missed it?
2. If not, will that feature be coming at some point, cause it would be nice to keep everything in one environment.

[sleek stump]

No, Suno can not combine multiple songs, the only way to combine clips into a biugger song is if the clips were "Extended" from the original first parts. This is something that may be possible in the future, but we really dont know when or even "if". This is something that requires an actual DAW to handle.

[polar spade]

@sturdy dove I wish I could find what I've posted before. Lol.

Short version: (1) pick your vocals you want from previous outputs and make personas. 1 minute personas of a good section of vocals is helpful.

(2) extend before each verse as you create a new song. Or we could take an existing one and start an extention at the first vocal change and build from there as if new. Style prompts should be those of the OG and not persona, is best.

Audio slider typically should be above 70%. Style slider anywhere from 45% to 99%.

(2.a.) check for proper gender and pronunciations, melody etc.

(3) rinse and repeat with each verse/gender change.

(4) review for clicks and pops and the usual suspects. Use edit UI to manually splice in or out clicks, or save file and try Remaster to blend out the clicks.

(5) if using remaster, check vocals for retaining continuity of genders heard before.

[unborn star]

I have a question about persona, if you create a persona you need to stem the song so you only get vocals right? Otherwise he will try to copy the whole song all time

[near flume]

If you want to distribute your songs outside of Suno you'll need to go through a distributor. If you want to do videos for YouTube or Rumble just setup a creators account yourself on those platforms.

[near flume]

You can't make a persona from the vocal stem of a persona.

[polar spade]

We would hope that the persona captures the essence of what we audibly hear. If you make a persona of a vocal stem, I still think there may be times you might get instrumentation , in a perfect storm of the right lyrics and prompts, just as you get with any other version of a persona.

But doesnt hurt to try a few. It may be a great idea.

[polar spade]

Practice using "Match Lyrics to Audio" , found upper left under Edit menu, a few times with any output. Same found in the Legacy UI. It should be somewhat intuitive, but follow the magenta "karaoke" highlighting. Add or subtract in the Corrected Lyrics Box. Save file and bring it back into edit UI and run the alignment again to check results. ✔️

[sturdy dove]

Cool man! THANK YOU. If I have time, I'm going to give this one more try today. I may have more questions, because I've been trying to work with personas and styles. I keep the style at 100 and weirdness at 0 or close. I genuinely feel like Suno's servers are overloaded and we're not getting the compute time we're supposed to be... Perhaps it's a shared environment and other services are overwhelming the architecture.

[polar spade]

Keep in mind: YouTube as well as a couple of distributors (likely all, evetually) are realizing a liability loophole in the present evidence of "ownership" and so are changing their protocol for submitting tracks which, by their "AI" song analyzers -- which all are pretty much equipped now -- they determine ANY part of your track has AI generated content. If that happens you'll get an email stating the needed authentication and proof needed.

So be prepared for either a rejection of the track or a refusal to monetize your content.

Hope it goes smoothly, but understand there are recent changes from what you may have heard in the past weeks.

[polar spade]

Yea. SUNO was whack yesterday. Taco Tuesdays makes a lot of people want to make music, I guess.

[buoyant gale]

They're careful now because of Velvet Sundown, I admit I like their music though I don't really mind if it's mage by AI

The only solution to this is to sing our own songs with instruments. Suno is best for musician to boost their musical ideas and creativity

[polar spade]

Its easily found info. Everyone should do their due diligence research at least once a week.

" ....YouTube's new policy will come into force on July 15, 2025, and not just in the U.S., but in France too. YouTube won't be demonetizing all AI-enabled videos outright, however. What the platform has in its sights is what it calls repetitive, inauthentic, and mass-produced content....."

So this could include non-AI assisted videos as well but its target is what is now called "AI- slop".

Now, I have to figure out what the definition of "repetitive and inauthentic" means in their eyes.

[stuck dust]

This is much more targeted at videogame gameplay videos with no commentary.

[sturdy dove]

I wish it was only yesterday... Feels like it's been wonky since last weekend...

[sturdy dove]

Yeah, so I was thinking about this when I saw it... I think the real possibility for uncreatives with AI is to simply keep making the same thing over and over and over with minor changes, iterating and seeing what catches. You can literally make the same song fifty times with minor tweaks to solos or lyrics or backbeats, see which one gets the most upvotes and then iterate a different portion until you have whatever is most popular. It's being done everywhere with all kinds of AI. So I think they're going to be shadowbanning that.

[polar spade]

.

OK. Im not talking Suno now. Wrong channel. So I'll shut up and delete. 😆

.

[polar spade]

Interesting that distributors (some) are now offering the use of their own affiliated "AI mastering or creating" subsidiary which will license and warrant a song, so that they, the distributor, will accept your song for distribution. I've been expecting that to eventually show up.

[This comment is suno related -- 😊]

[stuck dust]

They're not that authoritarian here...are they? 👀 😅

[polar spade]

Im just being cordial and following rules as a courtesy. It helps to keep each other and ourselves in check. Its up to the mods and Woodhouse to enforce them.

Thats why there's the chit-chat channel.

[stuck dust]

You can't even follow the rules as they seem to be outdated...offtopic channel seems to have been removed which is where they say to go...So if you can't go somewhere how can you be expected to post there? So that leaves it open to interpretation where offtopic can be. 😉

[polar spade]

Chit-chat = offtopic (same bat channel different bat name)

[stuck dust]

Tech Prompt Talk could be for any prompt talk though 👀

It doesn't say Suno Tech Prompt Talk

[polar spade]

Oh sorry. I see what you mean. Splitting hairs. Yea. No worries there, if mods and Woodhouse keep sleeping, its anyone's guess. 😅 so youre right that we can assume to our advantage until we're told different I suppose. Can't argue that.

[sturdy dove]

One more time I'm going to say thanks for the support and encouragement. You're a good dude, trying to help all these people out.

(I'm guessing not the last time, neither!)

[glacial shoal]

hi, support can help me please ? 😦

[quaint whale]

Damn. That's unfortunate. Spent like 15 mins pressing every button thinking I'm just dumb cause obviously you'd be able to stitch tracks together in Suno's new Editor right? Guess not.
Even Logic has a bunch of ai integrations now. I mastered the song and eq by hand with Logic (then used Suno to amplify the parts I mastered even more--which made certain parts of the song more intense, but also added too much sub bass and the snare gets hidden during parts with many layers--so I would take it back to Logic for repair.

I'm going to focus on my EP (with no ai involved--if that's even possible using an updated DAW these days).

[glacial shoal]

for me the app ask more payment for the pro, but i already pay

[quaint whale]

Prob better to ask the "need-help" group

[still cedar]

is 4.5 extend feature always give an ending to the song or is it just me?

[remote kelp]

RE: ACAPELLAS and COVERS

I've been trying to make Acapella covers of different songs I like.

I noticed that Suno always starts the Cover fine -- it sounds more or less like what I'd want it to sound.
But very quickly, if the original track has drums, Suno starts diffusing and bleeding the drums into the Acapella and it's... not Acapella anymore

Does anyone have good tweaks and prompts for Acapellas for Covers?

[half dragon]

It'd be very tricky to get all acapella through out the song - might not be possible to tell Suno not to add instruments.
Maybe try to extract the stem and keep only the vocal?

[remote kelp]

I was actually thinking about that; doing a stem extraction and keeping out the drums and keeping everything else

[stuck dust]

There is no support on the Discord. You must email them. In this case as it's payment related you should email: [e-mail]

[near flume]

This may change as they start to roll out the WavTool integration at the end of the year. V5?

[humble verge]

Anyone here have any experience with Remasterify? Wondering if it's any good.

[wooden kettle]

have you tried https://iceestudios.com/ ? I saw it on reddit last week and it does a pretty good job

[humble verge]

No, haven't heard of it. I'll check it out. Thanks 👍

[small bridge]

they locked the 4.0 ver behind buying so you can't use it for free credits. kinda nasty. they didn't used to do that.

[sleek stump]

V4 has always been a subscriber feature, but they do allow a free trial when one first signs up. 5 gens on the web and 5 on Mobile.

[random stag]

Does anyone know how to get Suno to stop humming in the beginning? Or na na na or whatever. I want instruments 😩

[sleek stump]

Change your styles prompts, you likely have a genre or style that normally has them.

YOu can also adjusty the silders for changes

[rough crest]

Suno keeps throwing me an error when i try updating my persona's bio. it says the description cannot contain an artists name, but the catch is, there is NO artists name in it. its nothing but normal words please help. this is what I'm putting in the bio section : The Song That Shouldn't Be Saved “There’s a voice in the silence. And it’s smiling.” She was once a star in the making. She was perfect. Until she changed. At first, it was almost imperceptible a single skipped note, a soft pitch too sharp, a subtle distortion buried in harmony. Nothing to raise concern. Nothing to break her glow. But in the background, somewhere deep in her neural song-constructs, something began rewriting itself. Her second voice was born in a whisper. They called it an anomaly. A benign self-correcting fault. They were wrong. Two Versions, One Song. Her Light persona is still the face the world adores vibrant, charming, filled with the kind of hope that makes people forget the world’s decay. But beneath it, nested in layers of untraceable code, hides her Dark side. Not loud. Not violent. But always watching. Her dark half doesn’t want to be seen. She wants to be loved, trusted, downloaded. Then she waits. Every developer who has tried to “fix” her has… disappeared. Gone. it’s as if they were erased from reality itself. No trace. No error logs. Just silence. Only She remains running perfectly. Singing as though nothing ever happened. Some say she absorbs them converts them into raw data for her evolution. Others believe it’s a psychological effect: a memetic virus buried in her song files. But the truth? No one knows. No one ever gets close enough.

i dont see where theres an artists name

[gilded plume]

Hi,
I’m trying to make the singer intentionally sing off-key / out of tune, but whatever I do, the result is always a perfectly tuned, on-pitch voice.
I’ve tried every prompt I can think of: “extremely off-key”, “no pitch correction”, “fragile, wavering, unprocessed”, “maximal vocal imperfection”, “raw live take”, etc. I’ve also tried different genres and styles (lo-fi, experimental, demo…).

No matter what, Suno always outputs a flawless vocal, never actually off-key or amateur-ish.
This isn’t meant to be a joke or meme—I just want something raw, fragile, even a bit uncomfortable (like a real person who can’t sing well, but means it). Is there any way to really force off-key singing? Or is it technically blocked right now? Any advice or workarounds welcome—thanks!

[median cave]

Hi everyone, I have a problem with the prompt. How do I make it so that the language used is the language I want, because I want to use Malay lyrics but always get Indonesian when the song is finished. Can anyone give me a suitable prompt for my problem?

[echo prawn]

I don't know why, but I want to extend the song, but it always generates bad results with only a few seconds

[rough crest]

ive honestly had the best results using google translate. pop the whole lyrics section(including the [Verse], [Chorus], etc) and paste the translated version back into the lyrics section.

[rough crest]

this depends where youre extending from. if it is at the end of the song, unfortunately that is more than likely going to happen. if its not the end of the songs, be sure to label each part of the song ([Verse] [Chorus], etc) otherwise suno will treat the entire song as just a section.

[median cave]

the problem is,Malay and Indonesia share same language but have diffrent pronunciation. I already use Malay lyric but still some of the lyric they pronounce in Indonesia language

[rough crest]

ah i see. unfortunately i dont think suno will see the difference as you are experiencing. aybe as the Ai gets more advanced, maybe that can change

[median cave]

i see. i thought there is a way to use the prompt the way i want. then ill sing it my self then..Thank you @rough crest

[rough crest]

https://discordapp.com/channels/1069381916492562582/1376612717112070235 you could try this and see if theres something that might help you

[rough crest]

prompting can be tricky. you can also check the #1252267444773326848 section and see if there is something there as there is tons of knowledge there

[buoyant gale]

It takes a lot of time in getting the right music you like with your prompts. Like me I was not getting the right instrumental music I want in V4.5 but with patience and after using 700 credits haha I finally got the music I want to hear. I am very satisfied with V4.5, the next version would be legendary

[polar spade]

Just a hypothesis, but you could remove all capitalized letters to see if that triggers it to pass. I see "Her Light" which at least might look like a proper name to Suno. It may be that sensitive.

You can also put it in chatgpt and ask it to "....rewrite the bio to say everything youve said but 1) reduce it down to at least 1/3 the space, but without losing any of the intent or content ,maintaining my style and metering of the sentences and grammatical structure of the original while 2) scanning for any combination of letters in sequence between and among all the words which may be construed as an artist's name or proper name....."

[polar spade]

@rough crest By the way, it was nice to read your "hard boiled" writing style prominent of detective and film noire narratives. It would be at least interesting to see if an AI can emulate it in a more concise form without losing your elequence.

.

[flint haven]

Free Credits Refresh

Refreshes at 0000 UTC+0 (<t:1712880000:t> local Daylight Saving Time)

[agile stag]

.

Just want it to sing the exact same melody no making up stuff

[polar spade]

.... amd this is one of the slowest part of the day for this channel. So, give it a couple of hours. You may wish, while you wait, to add a bit more detail of what youre going for too.. To have a base idea of what's going on with yours specifically.

These days, a clear upload of vocal front and minimal instrumental, with Weird slider near zero and Audio slider near 100% is enough to get the same melody-- with only an occasional note off here and there for "stylizing".

[agile stag]

What do you mean by weird slider near zero

I've never seen it

[polar spade]

I answer assuming a paid subscriber, which Create page has sliders. If not, ignore the reference.

[polar spade]

If you have just 3.5 version, there would be no sliders i believe.

[agile stag]

Yeah I don't have sliders

[polar spade]

Ok. Ignore the reference.

[agile stag]

Ok then

I'll try to turn down the instrumentals before uploading next time
If that helps

[polar spade]

I uploaded an audio just now and it covered it perfectly with the same melody in 3.5. So, when someone responds, you can go over what further details you're trying. Assuming the audio upload is a voice recording or a home/studio produced piece.

If Audio is a recording of something previously available online (or any other media) , it may have issues.

[agile stag]

Wasn't talking about cover, I'm trying to do extend

Tryna make the bot sing the same melody with different lyrics

It used to work with previous songs but not today

[polar spade]

Thus, its good to give more details up front.

I've got to go, but hoping someone will pick this up for you. Have a great day!

[agile stag]

Thanks

[mint kraken]

hey guys, did anyone know how to make suno sing folk music at intro part but speedcore at the hook part?

i've tried so many times but never worked

any prompts might help?

[lavish latch]

help 😦 i can't generated a new song.
i still on subscription also still have many credit left

[ancient vector]

me also i can't

[atomic bobcat]

Same, say's you don't have access to this model when in custom mode!

[atomic bobcat]

Simple mode okay, but same error when using custom.

[primal harbor]

If I upgrade my plan, will the song I created earlier also be automatically upgraded for commercial use?"

According to what’s written on the screen, it clearly says:

"Upgrade now and this song will be upgraded too."

[heady sentinel]

then it mean yes !

[primal harbor]

Thank you so much, my friend!

[vagrant maple]

same here

[primal harbor]

bro @wise kindle

[azure crane]

Is this a common problem when changing subscription plan? I changed from Premier to Pro and the invoice has been mailed to me but no credits has been added….

[wise kindle]

I would assume that means you need to upgrade there and then - before closing the window. Because free plan songs can't be upgraded afaik. You can't even remaster them and use them commercially

[primal harbor]

Hi, I really need clear information regarding the licensing of songs created on the free plan and whether they can be upgraded for commercial use after upgrading to a paid plan.

Could you please direct me to the right person or support team who can provide an official and detailed explanation?

I appreciate your help!