📚┃suno-school

[sleek stump]

if you dont like either hit remix again and get 2 more gens

[trim egret]

Yes, I’ve used the replace section feature before, but I was thinking of a section as like the full chorus

[patent sand]

no, it means a section of the track in a general time sense

but it can do both

[trim egret]

Can I literally replace one word? Because I need to do this in three choruses and keep it relatively consistent

[patent sand]

best to do it one at a time

literally change the one word as sunnie said

[trim egret]

Ok

[sleek stump]

Yep, best results are 1-4 words in a short window, that preserves the audio levels and the backing

[stuck dust]

You will waste your credits. This does not work
The entire section will change regardless of what you choose
It is called "Replace Section" it is intended for an entire section to be replaced...but ironically that doesn't work either

[gaunt plover]

I'm not sure if this is the place to ask this, but if I remix a friend's song and give it to them who has the rights for commercial use?

[patent sand]

If it was theirs to begin with, and you remixed it, they have the commercial rights

[gaunt plover]

Gotcha, I was just making sure cause I wasn't entirely sure on how it would work exactly

[sleek stump]

Not true, I use this method often, and just recently a few hours ago.

[stuck dust]

It doesn't work though. When it doesn't mess up the lyrics the audio is out of sync.

[sleek stump]

Again, not true.

That may be your experience but it has not been mine and many othrs.

[stuck dust]

Out of about 50 attempts only 1 vaguely worked but the audio still was unbalanced and it was a 10 second clip...but it also gave a voice that didn't work or fit in with the rest of the song

It doesn't have any context length sometimes

[patent sand]

so you replaced vocals into a non-vocal section?

[sleek stump]

Thats not been my experience. I generally get what I want within 2 gens

[stuck dust]

I'm using V4.5 and V4

Both have been malfunctoning since V4.5 came out

[sleek stump]

Aside from Metal/Rock, I use 4.5 99%

[stuck dust]

Even the keep duration will not even keep the duration the same

It always includes a whole extra section that I didn't specify

[sleek stump]

Again, not my experience.

[stuck dust]

Are you using top tier?

[sleek stump]

I limit my selection windows to 6-10 seconds.

[stuck dust]

Top tier subscription I mean

[sleek stump]

I am a subscriber yes

[stuck dust]

Do you have the highest tier subscription?

Or the lowest paid tier?

[sleek stump]

Theres no difference between Pro and Premier except for credits... 2500 vs 10k

[graceful condor]

Boa noite

[stuck dust]

Which one do you have though?

[sleek stump]

I am on an Anual Pro, but what does that have to do with anything?

[stuck dust]

I may have a suppressed account then. It does not function whatsoever

They claimed they fixed it in an email but it's even worse than it was at launch

Unless it's one of those things where the website version is garbage and the app works?

[sleek stump]

Web page has many features the mobile apps dont.

[patent sand]

that's not how they shadowban people

[sleek stump]

Ive had 1 issue with the web interface, was a couple weeks ago, had a corrupted cookie and it caused all kinds of weird head scratcher issues.. clearing my cache and cookies fixed it.

[desert vector]

does waiting for suno to load before pressing play make better song or no difference? ill play them soon as i can lol when theyre still loading

[patent sand]

should be no difference in terms of the quality, but I've had a few issues in the past with the track breaking if I played it during generation

it would just delete itself and refund me

[sleek stump]

It doesnt change the generation at all

[patent sand]

better to just wait

[stuck dust]

It 100% is not functional for me. There must be something wrong with the account then, not sure why

V4 used to be able to do it and now it can't do it either

[patent sand]

report in #🪲┃bugs-web and send an email

[alpine perch]

I have a question has anyone noticed the bug where it looks like you don't have many likes on your song but then when you refresh it it just goes back to normal it does it on the app to not just the website so I'm wondering is this a bug they're going to fix or what cuz it's kind of annoying just saying

[sleek stump]

This is a web service bug... Im pretty sure its been reported

[alpine perch]

@sleek stump yeah but the problem is it's on the app to now and I've tested it out it's on both

@sleek stump because one of my songs when I was looking on the app said it only had two likes and then when I refreshed it on both the website and the app went back to being 40 likes

[sleek stump]

Same thing happens with credits display, sometimes it doesnt show the correct amount, but if yo urefresh it does... its all part of teh same issue I believe

[alpine perch]

@sleek stump yeah I was getting this weird paranoia that someone was unliking my stuff to be honest and I'm somebody like did not like my song it went to 40 to 2 was playing with my heartstrings yo

[desert vector]

also do u ever get your bpm temporight

does like it ever sync to the actual tempo u ask for

i keep trying to work out my tempos

[sleek stump]

Suno largely ignores tempo direction.. It will use slow medium and fast tempo is about it currently.

[patent sand]

32 people are constantly unliking and reliking your song at the same time? no dude. it's just a web issue. report in #🪲┃bugs-web

[alpine perch]

Oh yeah I know that it was a joke I know it's a bug LOL

[stuck dust]

They do really need to fix this though. So it apparently it works for some users and not others? never heard of something like that

[sleek stump]

Ya I get it... some people need to track how their song is doing.. Myself I dont worry about the social aspect of Suno's front end.. its mostly manipulated by bots anyway.

[alpine perch]

@sleek stump yeah this is why I always check both so that way I know I'm paranoid that way so I always check my comments cuz there's always a reply that I'll miss so I know there's a lot of bugs so I can hang in there for that that's no big deal I've dealt with worse LOL this is nothing it just makes me a little paranoid I understand it's a bug now after testing it out three times

[compact stump]

So does discord support the embedded videos inside suno songs? I haven’t had much luck so far in the videos showing up

[sleek stump]

Only in certain channels

Gen chat does not nor does it allow gifs or other media

It does here... Example...https://suno.com/s/HbpOSJ2xKVrsk209

SHowcase channels also have the video

[compact stump]

Thanks, i made some sweet looping AI animations and added them as videos to some songs and they play inside the suno player but no on discord. https://suno.com/s/pCDJXCDSWIdMSS1G

[sleek stump]

Thats normal, aminated covers do NOT play in the animated videos

[stuck dust]

Wow...You can't even remove existing lyrics??? Thought I would try removing the lyrics with instrumental and start over. But it just generates gibberish instead of instrumentals

[patent sand]

You have to use Custom mode and disable "instrumental", then add instrumental information to the lyrics box as instructions

It's relatively silly

[ember summit]

Sorry I am ruining the flow here. Is there anotehr way to post this?

DAVSON was kind enough to offer suggestions to fix my problem but I'm burning up credits and getting nowhere. (I know that seems to be the way here, I get it).

I created a song with my lyrics and gave enough instructions to get close to what I had in my head for instrumental and melody. But, Suno generated a missing word in a chorus and changed "I am" to "I'm" in the bridge.

I changed lyrics in the original song files phonetically, and saved, and then in the remix Lyrics area, too, and that only partially fixed the problem.

1-Missed a "Me" in "Marry Me (fixed)

2-In Bridge I have a simple "I've been here, I am here, "

It keeps giving me I've been here, I'm here, ..."

Syllable count doesn't work.

I've tried "Eye am here" (still gives me "I'm here")

Tried "Eyeam here" and it yields "I here"

Any other ideas? Using remix with original song and just trying to fix lyrics. Sometimes gives me better instruments (adds a piano in one) but always changes the3 syllable "I am here" to 2 syllable "I here" or "I'm here"

[half dragon]

Maybe try "I-am-here" ??

[ember summit]

Will try. You ARE meaning with the dfashes, literally, right? If so I had not

Thanks.

DIdn't work. STILL creates "I'm here" which works better than "I here" but the 2 syllable "I'm here" just does not work as well as 3 syllable "I am here"

Grr. But good idea, so thanks! Going to sleep on it before spending more credits

Really flows better with:

I've been here
I AM here
and I'll be here when,
you sing me the song, you sang to me oh, way back when..

[bronze shale]

try line break? or a comma?
suno does like to eat up your words

[dusky salmon]

I emailed Suno about the billing issue several days. No feedback or help resolving my problem.

It is urgent since I cannot use Suno now.

[keen stream]

Hope a fix for the music quality drop after halv of the song will be fixed. Made some perfect song and the instruments sound like a blurry chans

[high hound]

Anyone know why Suno support are refusing to reply to me, regarding my account not being able to generate new songs, gives an error message, and I have asked them 3 times over 1 month, and never get a reply?

[hollow hare]

Billing and Support do not work on the weekends. You should receive a response from billing in a few working days.

So if you sent an email outside of Friday's business hours (in Eastern USA time), I suspect you should get a response by Wednesday the latest, based on previous reports by other users

[hollow hare]

Support has a HUGE backlog. One user reported a while ago that they got a response after a month.

Maybe we can help you out here? Can you describe what you have done and what errors you received?

Commonly, situation of this nature typically tends to be:

• Using copyrighted material. Not only is this against Suno TOS but also against the laws of many countries (and violation of international copyright treaty if your government signed on to it)
• Using public domain lyrics. This is against Suno TOS as it requires exclusive rights over the inputs. Since you cannot have such control over public domain materials, Suno has prevented the use of such lyrics
• Are you trying to make a persona or make a song public that was derived from an uploaded audio file that has vocal-like sounds present? Suno restricts the derivatives of such outputs from being made public and cannot have personas made sue to impersonation concerns. (Vocal-like sounds may include some unexpected entries such as didgeridoo )

[polar spade]

Great information!

[dusky salmon]

I am currently a Pro subscription and want to upgrade to Premiere (higher tier)

The payment on the desktop did not go through. I used the same card that I used two weeks ago.

Then, I changed to at least three cards and no success.

I went to the Suno app, it does not allow me to upgrade.

Now, I have no clue how to resolve this and submit a ticket to Suno. And as you said they are behind.

I think customer support is a very basic requirement. How could they run biz in this way?

[hollow hare]

Simply put, they have a lot of users and not much staff, both team and support wise. A team member did confirm that support get thousands of inquiries per week.

[dusky salmon]

Is that something they are solving?

How come I as a customer did not be informed ahead when asking for payment that Suno biz has no basic customer support?

If you want money from a customer, is this the way to run it?

[hollow hare]

Unfortunately I have no idea. It is something I did not asked because they can only disclose so much information to a person not part of their company.

[dusky salmon]

I am mad because this is fundamental and I did not expect to experience almost zero support in 2025 from an AI based company.

Ridiculous.

Always credit

[hollow hare]

Honestly, I never heard of companies disclose such information like that, ever.

Could be a lot worse: I just learned recently from Linus Tech Tips videos that HP instructed their customer service reps to not pick up the phone calls for at least 15 minutes so that many of the costumers would give up (thus saving the company some money? I honestly do not get their thought process)
https://arstechnica.com/gadgets/2025/02/misguided-hp-customer-support-approach-included-forced-15-minute-call-wait-times/

Oh right, they did it to force people to look into their problems online...

[polar spade]

I like when calling before open hours to a company and getting the "due to call volume higher than normal there will be a longer wait time...." recording instead of the "we are presently closed. Please call back during our normal hours" 🤔 Proves that's being implemented by that company.

[dusky salmon]

Slow burn of commit suicide.

Why don't we focus on Amazon which provides superior customer support?

Why are we paying attention to the negative biz which is so poor?

[blissful veldt]

Hello, I have more than 50 song pages and I would like to delete them all at once since I've kept my favorites. Is that possible?

[hollow hare]

In short, it is about money. If Suno wanted to focus on increasing customer experience with more support staff, they would likely need to:

• Increase the prices of the subscription to help pay for the increased staff
• Elimination of the free plan for the above point and to reduce the number of people who will send inquiries, which in turn reduces the number of extra staff needed

You are not limited to using Suno for music AI generations, there are a few competitors as well as several open sources where you can run your own Music AI (provided you have the GPU to run them).

Anyways, please see Dasvon's message. Did you check with your financial institution first to ensure that the rejection did not occur from their side?

[hollow hare]

I believe that you can do shift + left click on a PC to select multiple entries at once. Select an item first then scroll to the last item and shift+left click it

[blissful veldt]

Ah okay, thank you

[dusky salmon]

I am certain it is not from my card issue.

[hollow hare]

1. Issues with payment processing through the browser would be related to Stripe (payment processor) or your bank.
2. The failure to be able to upgrade from the Suno app could be attributed to an issue on Suno's end and/or Google/Apple, depending on how the system was setup and what the issue was. Or could be your bank
3. Check with your bank. Even if you do not think it is an issue on your end, you will want to confirm it and provide that as evidence when talking with billing/support

There can be several reasons why your cards are not being accepted. These include:

1. The bank blacklisted Suno due to high chargeback from other people or they considered your transaction to be suspicious
2. The cards were gift cards which many payment do not accept due to fraud.
3. Fraudulent cards used. If you are going through a middleman, this would likely be the case, especially if they charge you less than the actual subscription price
4. Geopolitical reasons (e.g. payments from sanctioned countries are likely to be declined)
5. Insufficient balance (you would be surprised how common this case is)

[blissful veldt]

I have a button 🚫 when I want to click on created there is an update in progress?

I thought it was because I had too many songs in my library, but even after deleting over 100 tracks, the 'Create' button is still blocked.

I still have 9,600 credits though

Apparently, it's the 'Create' button in the Custom section that's blocked — the one in the Simple section works fine

Ohhhhh sorry, apparently you have to choose a persona to be able to create in Custom — I think that’s new 😜

[paper gull]

Generally speaking the new 4.5 Sounds fantastic.. Kodos to the entire team on work really well done.
Creating and Extending the Music works really well..
Even extending and only just approximating the extension time, it works it out.
Just make sure that the Lyrics and Pronunciation (Spelling) is perfect first go, Because you CANNOT EDIT.
Which is a problem if you also make music in other languages
By well we mean, Almost really well.

Extension
Since Editing does NOT WORK, you can go right back to extend the song. Right back to the Intro and even almost to the beginning. However...
The Biggest Problem Extending a Song is that

1. You can only Extend by **ONE **Verse at a time, otherwise it will pick and choose it's own order and they are NOT the order you listed the Verses in, even if you numbered them correctly.
2. It often uses the originally created song words below, instead of your NEWLY inserted Verse. So you can often end up with a garbled mess.
3. Of course every generated new addition (Part) is a slightly different version of the music too, but generally you can end up with almost the same Version and rarely a better version. But you can while editing add additional [Control/Vibes/Harmonies/Instruments] etc...

Editing does NOT work at all...
The Editing features CANNOT

1. Stitch any change into the original song without MASSIVE abrupt edges, causing Cracks, Pops and Rhythm ISSUES, which are UN-Usable.,
2. Every change ends up with different volume instruments and/or even muffled vocals, so NOTHING you do will actually work in the original song.
3. The Replace function also VERY often add a longer section/block into your music and replace some existing, even if you asked it to stick the original time,
4. Replacing the Intro.. Very difficult., Honestly, it should be a function to cut the music and Extend by adding a NEW INTRO.

||We are so close to being perfect||.

THE EDIT FUNCTIONALITY NEEDS LOTS OF WORK..
So TOO does the Extend functionality.

We ACTUALLY Believe that THE OLD EDIT Function was MUCH BETTER.
CAN WE TRY THAT, for now?.. DO YOU HAVE A LINK to USE THE OLD EDIT FUNCTION?

[peak wasp]

it seems to me that the code for generating the wav need to be rebuilt from scratch. the distortion and high pitch scream that gets worse as the song progresses when making heavy metal songs is still just as bad in 4.5 its worse during the pre-chorus or chorus or ant part of a song that repeats. seems clear that it is trying to copy parts of the song and modify it but instead of replacing frequencies, its just layering it into the new part and adding distortion on already distorted frequencies. I have tried two other generators and it seems each one does this to a different degree. Riffusion seems to be almost as bad as suno and musicful actually seems better although the problem is still there to some degree.

[keen stream]

Worst time thieves are people who send stupid questions they could google and get an answer in 10 seconds. They should have ai that reads the mail and auto answer. But they got lot of money so they could fix lot if they want

[round bobcat]

Did anyone find a trick to remaster in 4.5 without changing the style of the song but get only rid of the shimmer? My problem is that 4.5 changes the stress of the words. If the singer's voice used to go down, it now goes up, which distorts the mood of the song.
When the voice goes down, the sung word sounds powerful, when it goes up, it sounds soft like a blubbering sissy.

[distant burrow]

Dear Suno team,

Please be aware that if a public persona is linked to a song marked as private, then any use of that public persona causes our own song to be automatically classified as a remix. This is problematic, if the original track is private, we have no knowledge of it, nor any access to it. Labeling our creation as a remix under these conditions is inaccurate and misleading.

I hope this issue can be addressed and resolved promptly.

[still otter]

Hello can we ask suno ai to make just a 40 second song or not? I have the premium version

[digital shoal]

Umm so i was charged for my monthly subscription, the renewal date updated. & yet my credits have yet to be reloaded…. Support doesnt say anything. So am i at a loss for money or do i need to sue suno over $30 😂

[neat nexus]

Have you tried logging out and back in again to see if this refreshes the credits listed on the account?

[sharp moth]

Song downloads are still broken

App works webpage don’t work just makes the song a example and removes the name and everything

[slender oracle]

if i try and extend an instrumentsl piece, wih an extended form that has lyrics, and the lyrics are really short, it generates a 2 second clip that causes a credits refund. every time.

i didnt do this on purpose, thought i was putting style description in the lyrics box because it became re-enabled when i went to remix

[patent sand]

You can try prompting a 40-second section to be something specific and then using Crop + Fade Out in editor

[patent sand]

If it is for sure past renewal date, email billing. support will take too long

[patent sand]

Some people have had issues downloading on mobile. Current workaround is browser download and then file transfer

[patent sand]

Try replace section if it's really short

[dim mantle]

I have over 3,000 credits but when im in the editor, it says you need more credits. Also when you guys updated the site to Remixing, theres 2 of the same words saying remix but you removed the reuse prompt which i was using till it was gone.

[patent sand]

Reuse prompt is now "Use Style and Lyrics". you can click off of Remix after that and it will look normal

[dim mantle]

Im still having an use with the editor. It says i have no more credits but i have 3,380

also, how do you use the same vocal voice. Theres a voice that works perfect for the songs im making

[patent sand]

Unsure because I'm on free plan. Maybe email billing to get it in the queue

[patent sand]

Stem split the vocal and remaster, then add to persona. It will keep the vocal

[dim mantle]

whats their email

[patent sand]

[e-mail]

[dim mantle]

thanks

[polar spade]

For jingles and Sync incidentals. ( @still otter )

"Jingle length" songs can be made fairly easily.

Prompts such as jingle, announcement, memo, TV theme

1. a No lyrics (output) on Customs mode can often just make short songs and hit near your :40 automatically by chance.

2. a short verse, like a slogan or jingle style (a few lines and a zinger) and Suno will typically keep it below 1 minute.

3. Spliced: Attempt the above (or just run a full 3 minute song, it may not matter) to get an output between 1 and 2:30 minutes. Then remove the middle to leave only the first couple of bars being your jingle length ( all around the mulberry bush .....) and a few seconds at the end for the zinger/ending ( ....pop goes the weasel).

Each of these you may remaster and/or cover to help remove any splice marks (pops, cracks,....) . And then can Cover to try varieties to suit.

The arranged is just a small example of taking a 1:13 song and cutting back to 20-ish seconds and one 7 second snippet alternatives.

https://suno.com/playlist/206a22a2-f7a8-463e-8836-1e63fed6f1ea

[polar spade]

See above. ☝️

[still otter]

Thank Dasvon I'm going try

[polar spade]

Regarding "stiching" .

Practicing by anyone to learn the skill of making as precise splice nodes as possible (exactly as one splices in DAW), then Remaster (or even Cover) to blend has taken care of 100% of any splice edits I've done.

Sorry to say, this may be beyond the efforts an average user might wish to take, which is understandable. But the option is there.

[dim mantle]

When i try to create a Persona, it keeps saying , "A Persona cannot be crated from this clip. Im using the vocal but does this every time.

[paper gull]

But as mentioned by many people, remastering or a new cover NEVER (or extremely Rarely) creates the same song.
Once you get the right inflections in the notes and especially the voice.. They are destroyed in every effort.
And trying even down to the second, there are still issues... ONE can get extremely lucky but mostly, it is impossible

[neat nexus]

Try upscaling\Mastering it then see if it works. Failing that try covering it and seeing if you can get the vocal to persist properly and then create the persona from the cover.

[polar spade]

I wrote jingles for a radio station while I was in college. So I still like to write a few here to test. But writing incidentals for movie/tv syncs is also important part of the music industry now. These can be done in Suno easily, keeping in mind that industry doesnt like AI at the moment. You find ways to work around it.

[gilded mortar]

Is there any feature that can combining multi songs that i created in one track?

[dim mantle]

In the editor when the Ai uses your credits and then ruins the Generation files, making its own mistakes, credits are wasted and used for its mistakes. We shouldnt have to waste credits when Ai ruins the lyrics

[polar spade]

Meh..... kind of.....

There is not a means if clicking on your favourite songs and clicking "Mash-Up" and getting one output. That's for version. 7. 😊 I've already suggested that. 😆

But you can try:

Making a Persona of a small snippet of each of your favourite parts of each output. Then make extensions using each of these personas until you have a full song.

When you now "write out" your prompt or click the "magic wand" this is basically what Suno is doing already for you, but not as distinctly as if you spliced persona after persona together.

[polar spade]

Though a valid concept, historically AI, since beginning its journey a few years back it is presumed to make "mistakes". We now never assume there will be 100% accuracy. It's just about how that's related to use and cost to the consumer that you and many have issue with.

I'm sure if unlimited, there would be fewer mentions of it.

So present in the suggestion channel your solution as to the levels of cost per levels of "mistake" and it could help.

As example: Many mention they should not be charged for 10 second outputs or 20 seconds, or 40 seconds, instead of the now 7 second or less. That's fine, if enough agree on the same duration, there could be a different cost structure implemented.

[sharp moth]

That’s the opposite of me apps working webpage page isn’t

[gleaming basin]

Does anyone know how to edit the payment method?

[polar spade]

In a browser, its clicking directly on your credits to bring up the subscription profile. Then click Edit Billing Details.

[gleaming basin]

Thank you

[raven turtle]

V4 Personas work beautifully... BUT... v4.5 is completely broken. It just copies the persona completely. Riffs, vocal lines... everything! Arrrgggghhhh

[trail fulcrum]

Hello, I'd like to contact the Suno team. How can I do this? I've been trying to cancel my plan for days, and it won't let me. I paid for a month. Since I'm having problems with my computer, I'd occasionally go into the month to use the credit I paid. I didn't think I had to cancel it. Because if I pay for a month and it's not the promo, then why does it automatically want to buy me another month's payment? I need to resolve this, but I don't know who to contact. I was only able to recover my computer today. The card I used to buy food. Now, I'm unemployed, and every day it wants to buy me a monthly payment. I have no income. Can anyone tell me where I should go? I'm sorry if there are any problems with the writing. I'm using a translator since I'm from Argentina.

[polar spade]

Clicking Profile/Avatar. Then click Subscription. You'll find "Cancel plan" . Click Cancel Plan once to cancel your present subscription.

[trail fulcrum]

the problem that does not allow you to cancel or cancel the card.

[polar spade]

If you have no funds in your bank account and it has been attempting to recharge your bank account, then that is an issue which you may need to be handled with your bank/credit card issuer (though not likely) or directly with [e-mail] (likely ) for them to contact Stripe to discontinue the attempts

[trail fulcrum]

Thank you very much. I'll send an email to see if they can help me. Thanks!

[polar spade]

If you click the Edit Billing Info it. Also has a Cancel Subscription directly through Stripe.

[polar spade]

Removing card info: correct. I dont think there is a way to completely remove the payment information without "deleting" your account. That is a fraud prevention step with many online companies. So "canceling" the payments from happening seems to be the choice, without speaking directly with Suno.

You can delete all payments registered EXCEPT the default payment.

[trail fulcrum]

The card is the default method. There's no option to withdraw a subscription from Mercago, and there's no way to do it from the app. So I ended up turning to support for help, but since it's in English, I didn't know what email or other way to contact them. My English is limited, and I'm using Google Translate to help. 😓

[daring lava]

🚨 New: SUNO HELP Category 🚨

We’ve added a SUNO HELP section for support-related questions.
Check the layout here: https://discord.com/channels/1069381916492562582/1372994443656957952
We’re also updating the Suno-FAQ here: https://discord.com/channels/1069381916492562582/1372994765196361774

If you think something’s missing from the FAQ, just let a mod know!

[trail fulcrum]

Thank you very much.

[patent sand]

What's the format for #archived-ticket-help? Without dropping my email for 28000 people to see. Have an ongoing support email (as in, back-and-forth replies) and not heard back since last Tuesday

[daring lava]

Yea definitely don't share any personal info there. We're leaving it pretty wide open for now with plans to better refine it along the way. Main thing is to have a channel dedicated to support-related things for members to be able bring up what they need to.

Letting us know that you haven't heard a response from a support team member in a reasonable amount of time sounds like a good thing to bring up in #archived-ticket-help

[half dragon]

Might be better if it has "thread" instead of "chat" format?

[daring lava]

Great suggestion! Thanks for the feedback 🙂

[patent sand]

Maybe also make a note to verify that any DMs are from someone in the server with the Team role. This hasn't happened yet, but someone could impersonate an admin to get emails or other details

^ if they do that via DM to begin with

[rose jasper]

What might be the issue? I got funds on it..
Buy More Credits

An error occurred.
We were unable to charge your saved payment information.

Oooh, figured it out, wrong default card nvm 🙂

[quick wedge]

Do you know if they're looking for a new mod team for the channel so it's not chaotic?

[neon vortex]

Question:

For reasons unknown - when I download one of my tracks to do some final touches in Audacity, then soundquality is complete dogsh*t, despite downloading it in .wav - and when I compare it to the playback from the website - it's night and day. This is the first time I have actually ran into this problem, this severely. Like - the sound is just absolutely trashed. What do? Q.Q

[patent sand]

You can just report the comment if you want

My point is that random comment beef doesn't belong in this channel

this channel is specifically for discussing model usage etc.

[patent sand]

Which model? 4.5?

[neon vortex]

Si. I know that the quality is sometimes somewhat questionable with 4.5 and the workaround is usually to just do a quick remaster, sometimes it works just remastering the same song to 4.5 again, sometimes I just go to 4.0

[minor sandal]

Last tuesday lol. I have something outstanding since December. They hope you just drop it instead of actually assisting.

Maybe if they supported their customers, the main chat wouldn't be?

[minor sandal]

Notice how your songs sound quieter? its an issue with the volume. You have to amplify it down, Suno is pumping out songs that have maxed volume to the point that they sound weaker. I know, it sucks

[daring lava]

Hey Jake! I'd like to figure out what's going on with your ticket. Going to ping you to #archived-ticket-help channel.

[neon vortex]

Yeah. The volume is way down most of the time but that is an easy fix. What's really causing me to scratch my head right now is why in the world the track sounds quite "okeyish" when played on Suno, via the browser - but once the .wav file is downloaded and opened in whatever program able to playback audio (Audacity, Media Player, VLC etc) - it just sounds like someone took parts of the track, shoved it into a concrete mixer, turned the mixer to full throttle and just shoved some rocks in it for good measure.

[eager talon]

Hihi o/ I was wondering why all the songs I’m currently making are repeating lyrics that have already transpired. I don’t have any callbacks to it or lyrics within the song that are making it loop back

It’s completely skipping verse 3 & the outro midway through verse 3 & looping all the way back to verse 2

[polar spade]

Since its out there, I want to say, personally, I'm sorry to hear of your mother passing away. Having recently had my mom die in 2020, I know there are many frustrating areas where hospitals and staff could have done better. Sorry again and my condolences.

[eager talon]

It eventually gets to the end just repeats the 2nd verse & beyond mid way through verse 3

I’m so sorry for your loss 😦 I didn’t see the messages beforehand

Thanks for the feedback!!

[sleek stump]

The model likely is trying to maintain what it thinks the structure of the style you chose should be. Many genre of music repeat last verses or bridges and chorus. This “may” be what’s happening.

[patent sand]

any chance you have two verses next to each other? I've had some luck in the past changing:
[Verse 1]
lyrics
[Verse 2]
lyrics

to just:
[Verse 1]
lyrics
lyrics

the AI seems to get confused when there are many verses

[eager talon]

I didn’t & I tried to break up the verse 2 & 3 by adding the chorus after the bridge in between which fixed it for more of verse 3 but still loops back to verse 2.

I did just purchase premium again & don’t seem to be encountering the issue but I haven’t tested much

[sage jetty]

seems like the lyrics when creating a video are messed up again.

[limpid heath]

Hi, I would need your help please: I would need to create songs as long as I want (well max 15 minutes), that sound like my already existing music. If I like one of the proposed songs made by Suno, I would like to be able to edit them, by explaining what I want in the prompt, or just making them longer, adding an intro or outro... Is this something I could do with the Standard Subscription?
Thank you

[slate trail]

What can we expect with v5

[half dragon]

Yes, it's possible; except the editing part, only Pro subscription has access to Edit tool.

[daring lava]

No official word on that yet, what would you like to see from it?

[slate trail]

Making proper duets with your personas. I created sister personas. Having the style of music from 1,000 characters to maybe say 3,000 would be nice

[royal sail]

My name is Fernando, and I'm developing an application that will be showcased at The Town Festival, a major event in Brazil.

I would like to know if you offer an API service that allows developers to purchase credits and integrate Suno's capabilities into their own applications. This feature is essential for my project, and I’m very interested in using your technology as part of the experience.

Could you please let me know if this is possible and how I could get access to your API?

Thank you in advance for your support. I’m looking forward to hearing from you.

Best regards,
Fernando Filho

[unique stone]

Suno do not provide or allow API access and at there are no current plans to offer it.

[patent sand]

Pretty sure that's a bot

See #archived-ticket-help. you'll know what I mean

[unique stone]

Yeah I just noticed that. After looking at the name discrepancy here.

[royal sail]

Hello people, sorry I wrote wrong my name, i'm brazilian girl and I put my boss name in the text

I'm not a bot

I'm creating an application for the festival The Town where people can choose 5 words to change the chorus of a song, and I was researching AI because Suno could help me, but I can't use it directly on their platform; I need to use an API for my application.

[patent sand]

Sorry for assuming then

[royal sail]

I wanted to know about an AI that could provide me with a service that makes music and offers an API.

[royal sail]

no proble

[patent sand]

I don't think Suno has an API

[royal sail]

problem

[patent sand]

There might be something open-source out there

Please be more specific

Declined payment? Delayed credits?

#support-guide, Billing section

Someone earlier had the same issue but just had the wrong card apparently

[deft parrot]

Alguém com dificuldade de subir áudio e conseguir trabalhar a música com ele ?

[torpid thicket]

SUNO links not working???

[sleek stump]

A little more information might help us help you...

[lyric knot]

I can't use the website to create songs. There is a pop-up window prompting a system error.

[polar spade]

I get them occassionally. At best, its just coming back later to see if there is no more error.

[ember summit]

Hey Dasvon, me again. Same problem.
(Using COver method, changed original words AND in new section of cover, too)

Used 80-90 tokens trying to change
"I've been here, I AM HERE, and I'll be here when..."

Keeps contracting to "I'm" , or, it deletes "am" going to "I here"

Tried:
eye am here
eyeam here
eye yam here
I-am-here

Should I try:

I
am
here
or
I, am, here?
Or??

[alpine perch]

I have a problem and I don't know what to do cuz I feel like I'm out of options and I don't want to come off as whining I know I do but this is actually something I can prove and also the fact they can even say this on here is disturbing on the suno platform

@sleek stump can I send you a photo of someone promoting self-harm from one of the chats I witnessed and no one's doing anything I'm surprised they were able to do this but yes I can prove it if you let me DM you

[daring lava]

Please feel free to send it my way. Which channel did you see this in? If you have a message link that would help incredibly

[alpine perch]

Thank you for looking at that David I don't mean to wine on here but I've never been so mad because to me that's disrespectful so thank you

[daring lava]

Happy to help! Thank you for reaching out!

[alpine perch]

I know I shouldn't let it get to me a lot of my friends tell me I should just you know move on but after seeing this behavior I had to report it there's no way I'm ignoring that that's unacceptable

[barren jetty]

I have a song with 2 voices 1 female and 1 male. I try to make them sing different parts and some in duet.
I try to overexplain where to sing with tags like [John] , [Anne] and [Both].
It actually works sometimes, but 80 % of the time it fails.
Are there any foolproof way to do it?

[quick wedge]

Duets are usually a hit or miss. In the style box you can add layered harmonies, male vocals, female vocals, male and female vocals, duet. Personally getting duets to follow your lyrics is tricky. I rarely create them. When I do create the duets usually appears on their own unprompted like one of the examples I have. In the lyrics box you can try [Verse - female vocals]

[barren jetty]

Thank you. Looks like it's 20 % hit and 80 % miss.
I tried both [Verse: female vocals] and [Verse: female singer], same same.
But when it's correct it's really beautiful.

[scarlet mica]

Alright. Neat. I'll try it

[quick wedge]

If you search around Suno there might be some duets around that you can look at for inspiration on how the style and lyric is set up.

[barren jetty]

https://suno.com/s/dXlVzK0hTXMs2Eov

[quick wedge]

Here’s one tutorial that someone else posted https://suno.com/s/2sgl5DRN01I147OG

You have a good song btw too 👍

[barren jetty]

Thank you. I will check it out.

[quick wedge]

You’re welcome 😊 If there’s anything else or any other questions never hesitate to ask and I’ll help you figure it out.

[barren jetty]

Thank you, I am really a novice when it comes to creating music, but i believe i am decent on lyrics.

[quick wedge]

You’re welcome. I’m kind of a novice at heart but I’m always learning new things every day. 50% ChatGPT usually helps me with lyric structure from my random ideas depending on what’s going on in my head. Most of what I create is varied genres. If you ever need motivation welcome to check my Suno profile that’s linked in my discord profile.

[barren jetty]

I checked it out.
I have much to learn.
I noticed you might be Thai.
Funny that the song i linked was about Thailand.
I am married to a Thai.

[quick wedge]

I was born there when I was younger. That’s exciting to meet Thai creators or people that’s traveled there before. I haven’t met that many Thai creators yet except for one person. It’s nice meeting you too 🙏

Welcome to the Suno community also 💬😊👋

[polar spade]

So, a failsafe approach? Keep clicking create on an extend until you get which gender you're prompting for, before going to the next verse. Rinse and repeat.

[buoyant egret]

#archived-ticket-help I am paid subscriber, recently I tried to renew my subscription but the site rejected my credit card, I need help so I can renew my subscription.

[lunar nexus]

Dasvon Hi! glad see you!
since yesterday, almost all the generation has errors. The errors are completely atypical inserts in another language that absolutely should not be present in the generation. I indicated this in the Error report, but there is no response. Has anyone noticed this?

and even these inserts have no conscious meaning - just a set of letters.

[pale raven]

Anyone here that might be able to answer some of my questions please?

[half scarab]

i could really use some help from a pro, anyone got a sec?

[hollow plover]

I made a playlist yesterday but it doesn't show when i look at playlists, only when i move songs.

[small parrot]

Many websites need to provide commercial authorization documents of AI platform or authorization letters of AI performers' timbre. Can your company provide them to solve this problem? Thank you.

[hazy sparrow]

👋 Hi Suno team, quick feedback/request:
I’ve noticed that when using different Personas, at least 50% or more of my generations tend to start with the exact same melodic or rhythmic pattern, especially in the intro and verse. Even when I try to vary prompts significantly, the songs often inherit the “base feel” or groove of the Persona’s origin track.

🎯 Would it be possible to make Personas more flexible or randomized in rhythm/melodic identity, so they don’t imprint so heavily on new generations? I’d love if the engine could preserve vibe and vocal feel, but allow for more variety in musical DNA (especially intros, chord movement, beat feel).

Thanks for all the amazing work — this platform is incredible and keeps evolving in powerful ways 🙌

[tulip depot]

This is a personal opinion, and not to hate on anyone who may disagree to me. I feel like the rights of the song suno gives you when downloading should NOT be locked behind a paywall. As I said, personal opinion, no hate to fellow suno members and it's community

[polar spade]

If you think your card is valid, one option is to try making the payment through Google Pay through the Suno app. That's a suggestion from Suno billing.

[flint haven]

SUNO BILLING HELP

For billing/payment issues, email: [e-mail]

Include:

• Your Suno @handle
• Description of the issue
• Transaction ID(s) or payment receipt(s)

Before mailing:

• Check if your card was charged
• Verify funds were available
• If Stripe rejected your card, try using the iOS or Android app to pay via Apple/Google Pay.

Do not send non-financial issues to this address — it may delay your response.

[polar spade]

These are in your lyrics? Just sets of letters?

Suno can "hallucinate" in the audio, that's just it using the voice as an instrument and babbling nonsense.

I've not seen suno to type text of random letters though. Definitely a new thing.

".....Can your company provide 'them'...." @small parrot

Please understand that 1) each website, company or artist may require different documentation. So no company can provide a specific one-size-fits-all document until its know what document is requested.
And 2) legally, many documentation (agreements, contracts, legal protocols, licenses, etc) ALREADY provides similar legal authorization without being a specific boilerplate looking form.

Read further for more information.

https://help.suno.com/en/articles/2746945

[polar spade]

☝️. Check the above post and read your TOS as your agreement and contract with Suno.

[pastel radish]

Can be a feature for self help. Just download the generic one for YouTube etc.

[polar spade]

I've got to say, I didn't follow what you meant. Sorry. What would one be downloading from YouTube?

[lunar trellis]

Hey everyone! I had posted in the general channel earlier in the week, but I created a small chrome extension to add “shuffle” functionality to the in browser playlist page. The extension was under review but was just approved:

https://chromewebstore.google.com/detail/pafceenjfanfeodomcaccmbeadndenlf?utm_source=item-share-cp

Wanted to post it here in case anyone wanted to try it out! If this type of thing is not allowed let me know and I can take it down.

Happy for any feedback / bug reports / feature requests! Full code is available here on GitHub

https://github.com/nateblaine/suno-shuffle

[pastel radish]

Generic copyright/licence documents for using songs on YouTube etc.. there could be a doc download button along with wave and MP3.

[polar spade]

Oh. Using a boilerplate form to fill in and attach as a commercial license for satisfying a requirement of proof for YouTubes requests. I see.

[pastel radish]

Yeah man!!

[polar spade]

Put that in the suggestion channel , give them some ideas. https://discord.com/channels/1069381916492562582/1260263379243761776

For anyone who hasn't noticed, there is now a "I need help" channel to help with questions and issues with Suno accounts and Billing. Check it out. https://discord.com/channels/1069381916492562582/1372994553233276970

This will help clear up this channel for its originally intended purpose of discussing how-tos and techniques for creating Suno songs or promoting techniques or marketing.

[tepid moth]

Anyone have any experience with the paid sub on Fadr according to extracting stems? I use the free plan and gets the stems out on mp3. I have an impression that it does not matter if I get them out as wav file. Any other suggestions (free) for separating stems (Not only the vocals)?

[polar spade]

You may wish to ask in reddit channels as well, or at least pop over to the General channel or. https://discord.com/channels/1069381916492562582/1159929934270181437 and ask in there as well. All channels are intended for AI related topics though, but those are somewhat more lax in that area.

[polar spade]

For the most part, I feel people have a diverse opinion on splitting stems and further, they have differing techniques for their use, typically being in a DAW. So it becomes a debate on which is best rather than any decent technique chat. I just leave it at that and dont go further, tbh.

[tepid moth]

Yes I know! Just got a very cool synth arp that I wanted to extract and wondered if if was possible getting it any better than the Fadr`s. Thats all. Anyways thanks for replying.

[polar spade]

As you mentioned, It's just separation of the vocal frequencies with Suno presently into a separate track along with the remaining instrumental. At the least, you could start with the instrumental track and download that to use in another app to split out.

[tepid moth]

Yes thats the approach I am using for this👍

[vivid wasp]

Any hints come out at what they may be trying to improve or add in v5 when it comes out next? 4 and 4.5 have been amazing

[radiant breach]

why do I have so many issues with suno? grr.... now I have a 4.5 persona that doesnt stay! Tell me why this first gen is used as a persona and gives the second gen? I just don't get it, every gen with this (first song) persona does NOT sound anything like the original. And any other persona on 4.5 is the same, they NEVER sound like the first one. WHY??

Persona -> https://suno.com/s/Jk4lsXUBQERXAAxH

Song 8/329 that doesn't follow the persona -> https://suno.com/s/qkTjRXx3V0FPrL1c

wasting credits yet again suno, fix your stuff

maybe the tiktokers you hire can get behind a desk too? have them see if they can fix the backend stuff.

another one nothing like the OG -> https://suno.com/s/YEfuHXJY9gcsbZoY

and If i make a persona out of this one ^ then I BET none of the next gens will sound like it. I'm done

put female in the prompt for this one -> https://suno.com/s/MMXKqpvwaMVkDyMh
STILL not getting the persona LMFAOOOO

<@&1108435550475653161> I'd like my credits back from this

https://tenor.com/view/naza-bart-gif-18252699006249252039

[maiden sonnet]

Hi ined help with my accaunt .

[lime wasp]

You can't really split out vocals by frequency if there is any sort of accompaniment.

[radiant breach]

I needed to come back because none of my personas are working

I can't get anything to sound like what it used to. WHAT HAPPENED? Who is flipping the wrong switches again

[unreal quail]

Make sure you save your creations, Suno will delete songs they don't agree with

[radiant breach]

suno personas are broken, who is on here from the company that can actually help? lol, this chat seems pointless

[half dragon]

You wouldn't even be able to create it if that's the case.

[neat nexus]

If you need that level of support you are encouraged to send email to [e-mail]

Meanwhile we're happy to offer suggestions but the moderators and guides are peer support and not member of the Team or suno itself.

[unreal quail]

Did and was in my playlist for a while

[radiant breach]

I'll just come back tonight and see if its more stable

[unreal quail]

Playlist caps at 50 without messing up the app. 49 this morning and found my favorite song gone

[neat nexus]

https://tenor.com/view/rick-and-morty-jerry-im-leaving-flounce-ok-then-gif-17149939

And honestly, it's a valid technique which I do myself.

[polar spade]

Thereby often having a bleedover. Right. The 90% success is then labeled "vocals" for the best practical description. It's a work in progress. But the process is to separate by frequency, nonetheless. In this case when speaking towards which tones or instruments, I was saying its the vocals only, which was the focus of the conversation.

[radiant breach]

I'll do that from now on knowing there's no help here

[unreal quail]

This Discord sir

Not the company

[neat nexus]

We can advise but are not developers. But we are also fellow users and the best support I've ever seen for doing this stuff is from asking my fellow users how to do something.

[radiant breach]

The Suno discord, but ok

[neat nexus]

It's AI.

¯_(ツ)_/¯

[radiant breach]

https://tenor.com/view/out-disappear-bye-vanished-gif-4932063

[unreal quail]

Im upset about my morning jam being gone. Really got the blood pumping

So I guess German motherland songs aren't allowed on suno? Seems racist to me 🤨

[half dragon]

bold assumption

[unreal quail]

Maybe I should of translated the lyrics Mmmmmm

[polar spade]

What's up , Woodhouse, with memes in the technical channel? Thought we had guidelines.😆

[unreal quail]

What is Discord without memes. But has anyone else experienced songs getting deleted?

[polar spade]

That seemed a thing when 4.5 first emerged, but I haven't heard mention in a couple of weeks at least.

[unreal quail]

I wrote it for my German coworker who had to return home. Just made him a jam about the motherland calling him home. It was....oh I see where I went wrong now that I type it out....

[polar spade]

Its 4.5 beta too. We should mention an issue to be reviewed and move on. Then share what DOES work.

Ill say, I did hope that, same as for 4.0 and 4.5 Remaster, there would have remained a, kind of, "4.0 Persona" creator in the menu, as well. Darn it.

[unreal quail]

What I do when I need to reset a persona is to find the original seed and crop the part I really like, then make a persona from that. Should be better with 4.5

4.5 persona's seem better as well overall

[polar spade]

I've layered a few personas this way until I got something back similar to the 4.0 version that I like.

Frankly, I think it's somewhat already becoming the standard if not obvious steps to upfit the 4.0 personas we like.

[neat nexus]

And this right here is the type of support that we have on this Discord. Peer support.

The team might be able tp speculate on when some feature or another will be altered (bearing in mind that this is a beta and things are still in flux) and give some sympathy.

But peers here will take the issue and dissect it with daily experience and offer solutions which worked for them now so the user can get back to making their song.

[somber wigeon]

helo

How can I upload audio longer than 2 minutes?

[ember summit]

I am new and tried to do the same thing a couple weeks ago and found out the system to date only will accept up to two minute downloads.

[polar spade]

The upload limit is 120 seconds for paid. So, specific to Suno , that's it for now.

[ember summit]

Hey Dasvon, me again. Same problem.
(Using Cover method, changed original words AND in new section of cover, too)

Used 110 tokens trying to change
" I AM HERE..."

Keeps contracting to "I'm" , or, it deletes "am" going to "I here"

Tried:
eye am here
eyeam here
eye yam here
I-am-here
I, AM, here

any other ideas?

[half dragon]

One more idea: try to break them in separate lines?
I
am here
or
I
am
here
See what it sounds like...

[tame tulip]

Greetings.
I am experiencing an issue with payment.

I accidentally paid for a subscription worth 96 dollars and would like to request a refund.
In suno AI
I need help

[hard beacon]

[e-mail]

[polar spade]

I got nothin'. What are they sounding like anyway? Sounds like Sunos stuck on an original rendition somewhere so it ignores what you're trying to say.

You could also change the meter slightly, by moving that line up to complete the line before, or move the line after to complete that line. Suno will reevaluate the meter and sing everything on the line slightly different.

If you want me
I am here
Just call me

To:

If you want me I am here
Oooo
just call me

Or:

     If you want me 
     I amm here -- just call me
     Yeaaa

[hard beacon]

Email: [e-mail],
Subject: [Billing] Your short issue summary,
Include:
Your Suno @handle,
A short description of the billing issue,
Transaction ID or payment receipt,

One issue = one email = one ticket = faster help.,
Sending multiple emails about the same issue slows down the entire queue and will delay your own response.

[ember summit]

It's doing various things. It seems to be mostly contracting it to "I'm" or eliminating the word am totally and saying "I here"

Now I am going back and making changes to both the original song that the cover is pulling from then the new cover where it asked for lyrics too. You suggested that earlier, cause it pulls from the original source when you do a cover.

[daring lava]

Hey there! Support-related questions have moved to this channel #archived-ticket-help 🙂 Regarding reaching out to support the steps mentioned here should help #support-guide

[polar spade]

Gotcha. OK. Yea, sounds like its stuck on whatever in the buffer from the first run or original.

Try the "all on the same line" and hope its gets it AND sounds right.

[polar spade]

Meh... could work. From my experience it can sing it in beat and it works, or it will pause between each word -- pauses might give a unique sound and you like it anyway, but its up to you.

While you're at it, sometime write a 1 word per line song, either it will rap it, or sound robotic and just say a word every couple of beats, but fun to just knock one of those out.

[daring lava]

S/O to all the people here helping others learn how to use Suno tools, mastering prompt crafting, inspiring others to create! It’s a big deal what you all do 🫡

[ember summit]

Much thanks to all

While that is remixing, and I'll let you know if it works but think I'll stop eating credits on it as I am really trying to line up a studio version of it anyway)

I took a song with voice ( created elsewhere before I became a Suno member) and fed in the first two minutes then told it to do a cover, but I selected "extend" from the one minute and 36 second point so it would pick up right at the second VERSE and I pasted it in the lyrics for the second verse, chorus, bridge, and everything else. What I got back was amazing, except, it did not include the original one minute and 36 seconds. What did I do wrong?

The other song did remix wrong again. Just letting you know

I had moved the troublesome lyrics to separate lines and all it did was once again eliminate the word "am"
I
aam
here

I am now trying to move
all up to line above to try to force it to re- process

[rocky root]

Gm guys!

Is it possible to upload my voice sample to use it as a base for a prompt?

[alpine perch]

Unfortunately no because of fear of someone using a popular person voice and copyright

[daring lava]

Hey there! Absolutely you can use your voice as a reference for song generations. Check out this article when you get a chance: https://help.suno.com/en/articles/2477633

[rocky root]

Arigato gozaimasu

[alpine perch]

You can use it you just can't publish it

[rocky root]

Wierd
Is it Premium or Pro?
I'm on basic

[daring lava]

Here's a good breakdown of what you can/can't do with a basic plan: https://discord.com/channels/1069381916492562582/1373496852112212121

[rocky root]

Found it in the library
Much love to everyone ❤️

[daring lava]

Not quite sure what you mean, could you explain a bit more?

[polar spade]

In general, you can't control any voice to manipulate it for singing specific melody lines.

There's no secret that it may pick up on your vocal qualities and at times get something quite similar. It's done that since uploads were available. But more rare than predictable.

Again, we can't make a persona from any uploaded vocal.

[rocky root]

I'm just hitting the wall:

I need a temple choir, singing in particular way

Like Christian choirs do, but somehow, it doesn't understand, so i thought if I'll "feed" it with my voice, to give it an idea, it could follow.

Sort of Enigma, Era, Georgian type of the music
Cant really understand how to give it a right direction

So....
Confused.
That's why i decided to come over for a help

[polar spade]

Oh. Style? Sure. Just upload a choir singing. See if it picks it up and disbursed in the song. Even upload the actual chorus, if that helps.

Try "boys choir" that always does a choir sound, but higher octaves. "GOSPEL" is also a good one, and include "ensemble" which is good for Broadway, gospel, rock anthems, audiences, etc.

[rocky root]

I'll get back with resaults in a few days.
Thanks a lot, Dasvon!

[polar spade]

You mean Enya? I would guess.

Add Chant as Gregorian Chant..... and you might like Native American or isicathamiya (African call and response chant)

[daring lava]

Interesting idea 🤔 Kind of like an exclusive channel where members that have demonstrated enough time and expertise can share prompt-related things? Am I understanding this correctly ? Trying to wrap my head around this.

[rocky root]

Marking as unread

[hard beacon]

for me it sounds like the content of the "Workflow" channel @tribal remnant

ok... so if you have to explain it in some words.. what should be the content of "Technical-Discussion" ??? Then the other question: You are talking about a channel were "expert users" can share their prompts experience for the best result they got?

[quick wedge]

Welcome to the Suno community 💬😊👋 Have a wonderful day too

[hard beacon]

ok.. so you made the experience that the "technical discussion" is more a "god damn.. muffled sounds... bad quality... what can i do?" context ? btw: just want to understand your thoughts 🙂

ok.. i think i get it 🙂

[polar spade]

Heres a couple pf example which may be in line with your styles. View prompts for styles used. Isacanthamiya is a very strong style and can be overpowering, so it takes some experimenting.

https://suno.com/s/Fcy6SXtMhFakOU2n

https://suno.com/s/Za5VryW6WNoMkcwY

https://suno.com/s/shr51pJIwb1VmtPu

[oak scroll]

Doesn't the audio function anymore? I made a song, want to use one as a cover and try to upload my older song, as I've done before, but it only gets uploaded and not "part" of my song, if that makes sense? (The same way as if I were singing the audio). Worked before (yesterday) doesn't work today?

[hard beacon]

but to be honest... do you really think people care about the specific channel topics??? 😄

@tribal remnant

ok.. but when i got a question but i am not able to write my question, how should i get my answer about my problem? Then it's more like a knowledge channel... if there is a solution about my problem - good.. but.. well.. you know 🙂

yesterday there was a guy in here with problem, that he wasn't able to login on the desktop version coz suno told him "phone number is already" used... he was a pro plan user on the app and want's to login on the computer.. but he didn't get it... in this case i was able to tell him the solution coz i had the same problem... and THIS is topic, that isn't shown in any FAQ or knowledge channel.. so there has to be a "user is helping user" channel... but.. how many channels should be created ?? 10 ? 🙂

[polar spade]

It would be fun if Discord had a mod control to just click and transfer a post. click "Your question has now been moved to the How-To channel for accurancy"
🙂

[hard beacon]

so... "prompt discussion" and "technical-questions" 🙂

or "technical-questions (user to user support)" 🙂

[polar spade]

I think I would keep getting moved to the "Wine and Cheese"* channel. 🤣

• << Whine and Jeez >>

[hard beacon]

From a psychological point of view, however, reducing frustration is important 🙂 If people can't find a suitable channel, then they will either use the “general-suno” channel or an open tech channel anyway. You probably won't find a solution for that in life 🙂

At this point, however, all users must also take responsibility for pointing out that there are channels for this 🙂

Do you think our conversation is a technical discussion, for example? We're in the wrong place here too 🙂

Hmm.. that's a good point 😄

I'm not disagreeing with you or arguing about the meaning... I see the whole thing as a relaxed conversation and just wanted to understand your wish 🙂

But if you want, I can also tell you how unhappy I am with the v4.5 result right now... but... oh yes... it's the wrong channel here 🙂

[polar spade]

Ill add to your idea the point that a bug channel is to POST bugs, only. That's it. They make a list and resolve the priorities. Then its moderated to remove posts that's aren't bugs. No help, no responses.

[hard beacon]

but then i'm just one of thousands of people who complain... i want immediate feedback that i'm soooooo right 😉 <--- irony

[polar spade]

I'd agree there's a balance both for commeraderie and marketing, to just let people vent a bit and not corral everyone into specific channels all the time. After 1.5 years in here, no one has ever burned the room down with their personal issues. No worries.

For the most part, especially in the general room, the peers will politely bring down someone who's just pushing buttons and trolling.

[hard beacon]

i know bro 🙂

That was also more of a sarcastic message 🙂

even if there was a spark of truth in it 🙂

btw: 1 to 1 translation from a german "use of speech".. don't know if it makes sense in your language 🙂

[polar spade]

Sorry. I wasnt suggesting you support my idea, many like to pop in there and help. Even I have done the same, until I realize its just enabling. So its become the same as this channel, and become a silt of personal issues to weed through to find the needed information.

If some realize its a No Response channel they will post elsewhere and the bugs can be tallied and reviewed better so we could be at 6.5 by now. 😁

Yes. We seem to be mature and adult enough to see them through, so I dont mind. I have my off days as well.

[hard beacon]

I think I read it somewhere earlier; I also think that some questions simply remain unanswered.

There is, and this is nothing new, the problem, especially with V4, that the quality deteriorates over the course of the track. At some point, it becomes very metallic.

But to date, NO ONE has explained whether the cause of the problem is known and whether a solution has been found.

I think that's exactly what could be optimized.

[smoky ore]

I have a song i absolutely love, but the Intro (first Hook in that case) is a mess. it's supposed to be female vocals, but it's not. The thing is, when i try to replace it, it turns out male again. I figured out that Suno works with the previous voice when replacing a part. So i really don't know how to fix it. Any ideas?

[hard beacon]

If you ask for feedback, then in my opinion you should get feedback.

and no.. I'm not swearing 🙂

I'm still amazed at what AI can generate in such a short time 🙂

[polar spade]

Putting [female vocal] as my first prompt in style prompt works good (along with the usual metatags to schmooz the correct gender) to get things started. Then I may end up having to extend for every stanza if I'm changing genders, but something i expect to do.

[daring lava]

Sorry, had to step away for a second. Just caught up on things here...

Would like to spotlight that all of us mods/guides are very open to feedback on server improvement, so please never hold back on that. This channel as is, is intended to be one where community members can ask questions about suno tools, prompt guidance, tips & tricks and so on. Really kind of playing it loose to see what vibe fits the best for you all. We've created a channel to cover support-related things to prevent that weight from falling on the shoulders of members here. Being it just started yesterday, expecting it to take a little time and redirection, but we'll get it there and I can't express enough how grateful we are for your patience and for all the instances of help you all have provided in this channel regarding such things.

Regarding the suggestion brought up, very interesting idea. If you'd like to put together something a bit more detailed about it (maybe in doc format), I can pass it along to the team for review for sure.

[smoky ore]

Duets work very well for me. 1 out of 500 is not the way i want them but in this particular situation every Hook is female except for the first one. Since the song is exactly what i want i have to fix that hahaha

[amber stirrup]

Have you discovered and chord progression or BPM directives that affect Suno? I’d love to see a chord progression header that works! BTW I’ve found that phrases like Dorian Mode or Gypsy in the song style help a little bit with adding non-mainstream chord progressions

[hard beacon]

There is a lot of feedback and questions in all channels. Maybe we should collect the serious questions and publish a weekly video in the style of "Q&A" 🙂 That would be user-friendly.

And then you have to be honest... if there's a problem for which you don't have a solution, then you should just say so 🙂

[smoky ore]

if it was any other part - no problem...but the first hook hahaha - right at the beginning of the song

[polar spade]

@tribal remnant Bam! There ya go. ☝️ You got an ear! Wtg

[smoky ore]

and i still haven't managed to have working personas 😄

[polar spade]

You can try to use Replace as well, if the song is already completed. Replace will take prompts.

[smoky ore]

that's what I'm doing but Suno won't give me female vocals. just one hook is messed up - the Intro Hook

[polar spade]

When usingReplace, use the style box and only put in "feminine" styles, they should NOT affect the overall style of the song..... so..... female vocal, beautyshop, femme folk, broadway, opera, k-pop, girl power, empowered.... whatever you can think of.

[hard beacon]

@smoky ore what model do you use to create your songs? just asking because: my personal experience is, that v4.5 works fine for duets, but the output isn't that good.. so.. create it in 4.5.. if you like the song - save it as persona.. then create a v4 song with that persona and the same vocals 🙂

[smoky ore]

my personas never work - i don't know why - they always put the old lyrics into my new songs. 😄

[graceful condor]

My loop I uploaded isn't being displayed in the create section.

[smoky ore]

yeah, that's what I'm doing at the moment haha, but i doesn't really work

4.5 btw and actually happy with it

[polar spade]

Ok. My hail Mary. When editing the present output that you like, go back and remove lyrics from the previous output and save. Then continue editing the present output with all your tricks. See what happens.

[smoky ore]

do you use the whole song or just parts to create a persona?

[polar spade]

Pull it up from the Library?

[hard beacon]

😄

[polar spade]

You asking me?

[smoky ore]

all of u 🙂

[graceful condor]

Yes, it is there but it used to have the wav form above the instrumental button.

[hard beacon]

i create it for the whole song in v4.5, then save it to use in v4 to get a good output 😉

[polar spade]

I have different voices in my head. This could be a problem.

[hard beacon]

https://tenor.com/view/evil-laugh-laugh-laughing-willem-dafoe-evil-gif-4145161

<@&1372996242325508176>: you fell like this ? 😄

[polar spade]

I like cutting songs down to pertinent sections before making a persona. Either a section or part of no less than 1 minute, or a song that's spliced up in a hodgepodge of all melody lines, riffs, and fills to help cover all stanzas and changes.

[smoky ore]

even with Duets?

[crystal prairie]

hi, are you still experiencing this?

[hard beacon]

yapp.. hold on.. i am just creating a track from a persona from 4.5 to 4..

[smoky ore]

means i could delete the female vocals and just use the male ones to get a persona

[hard beacon]

btw.. what kind of style we are talking about? @smoky ore

[graceful condor]

yes

[polar spade]

Nailed it!

[polar spade]

There's a thought. 🙂 I've pulled an old man voice ( about 5 seconds) out of a crowd noise sing-along once and personaed it and got a similar vibe of an old man voice.

[smoky ore]

okay a new editor just popped up - now i'm confused 😄

[hard beacon]

@smoky ore : what kind of style are we talking about when we are talking about duets?

[smoky ore]

Hip hop and Rap in this case 😄

[graceful condor]

I don't think it's using my loop.

[polar spade]

Alright. More toys!

[hard beacon]

ok.. i will send you an private message.. first the v4.5 track (not happy with the sound output but with the track idea) and then the persona created v4 version.. btw: maybe i also have an rap track with that workflow

@smoky ore

[crystal prairie]

post or dm me a link to your uploaded audio?

[crystal prairie]

cc @amber maple

[polar spade]

There is a Legacy Editor tab to the left if you want the old editor.

[smoky ore]

i kinda need to fiure out how the new Editor works 😄 just because i want to 😄

[smoky ore]

sure, thanks 🙂

[polar spade]

I see an "add section" and "rearrange song" . Might be what you need.

[smoky ore]

omg, too much new things...i'm lost

[dusky salmon]

Until now, still no feedback from Suno.

I have no way to fix this technical issue on my end. My project has been ruined from this matter. And I have no clue when support will help me. This is such a nightmare.

[graceful condor]

this is second track it didn't seem to use my upload, my first upload I tried was 1:28.

[smoky ore]

how to crop or delete using the new editor? am i blind? 😄

[hard beacon]

well.. the new editor looks lit.. but like Rebel77777 i have to figure out the functions 😄

[smoky ore]

😄

[hard beacon]

and i am pro DAW user.. but... where the hell is the Stem Button ? 😄

[smoky ore]

hahaha and how do i change back from the legacy editor?

[hard beacon]

click on legacy editor on the upper left 🙂

extract stems = 50 Credits ??? uffffffffffff

[polar spade]

can't yet do that Have to come back in from the new output

[smoky ore]

got it 😄

[hard beacon]

btw Rebel.. DM sended 🙂

[smoky ore]

how do i delete new sections? or is there only this undo thing? 😄

[hard beacon]

i am afraid of clicking the stem button for 50 credits coz i don't know the quality of stems 😄

[smoky ore]

I've tried it once and didn't like it...a week ago maybe

[polar spade]

Correct. Be careful. And for now, dont click the "voice and instrumental " stems.... its bugged

[hard beacon]

ok... then i will use the Create Stem Function in FL Studio 😄

At the point where the stems are frequency-independent and can be downloaded completely and individually, I will use it 🙂

[polar spade]

Well..... that's for V & Inst. The new stems are pretty nice. Less to no overbleed.

[unique stone]

In the new editor - right-click the section and you get an option to delete

[hard beacon]

sorry Dasvon.. don't understand what you wanna say 🙂

[lunar mesa]

vocal and inst should be working properly now 🙂 (recommend using them from the create page)

[polar spade]

Nope. Not. I just tried 2 minutes ago. Got burned 50 credits for double gens my mistake. 10 credits for double gens. So. Yes. Its clear to go. Sorry. Just double gens, so no worries there.

[hard beacon]

But the stems aren't individual tracks that are independent of the others, are they? So... what I mean is that the frequencies of the instruments are also the same as the vocals... if I create stems in FL Studio and deactivate the vocals, for example, the instrument track also sounds faulty.

I don't assume that stem extraction can be compared to audio tracks as you know them from the studio?

[polar spade]

Weird, but no harm no foul. Just delete two of them. Haaha

[lunar mesa]

i'll send some example stems in a sec for y'all to get an idea

[smoky ore]

not sure, if i like the new editor 😄

[steady kelp]

new stemming... different process. essentially, no artifacts 😎

[polar spade]

So my credits should go to 545. Testing once more....

[hard beacon]

If stems could be treated like individual audio tracks, that would be a game changer.

[smoky ore]

ok, this is not working for me...

[hard beacon]

@lunar mesa : can we get the full track of the stems for a reference?

[lunar mesa]

should be in there!

[hard beacon]

oh.. igd_REF... 🙂

just saw it

to be honest... it's a jazz style song.. an example of a more complex genre like pop punk or metal would be nice..

[polar spade]

Ok. All green. Debit 10 credits again. Delete two outputs. Its like a fresh day in Spring.

[lunar mesa]

i'll whip up a new one for ya

[hard beacon]

thanks 🙂

and NOW i have to say: that's the support and feedback i love 😄

[unique stone]

The models don't generate from stems (because they're not trained on stems in the first place) so they can't offer true stems and probably won't for quite a long time.

Existing stem splitters (including Suno's old one) would kind of "cut" around the instruments and vocal, which is a little destructive and/or can have obvious bleed through or dropouts.

This new one seems similar to something I had been misusing "Cover" for - i.e. it uses the audio as a reference and then generates clips that are focused on specific kinds of instrument. So it's generative rather than "subtractive" - there may be some bleed through between stems and it may miss things but overall quality is pretty good from what I tested.

[hard beacon]

imagine the output of the last hook output of a guitar sucks and you can replace it in a DAW from the first hook... holy 🙂

[verbal sun]

Are we getting API at some point or should I let this dream die? ||Okay either way, just wanted to manage expectations||

[shadow pier]

Here's a more electronic example

[lunar mesa]

something more pop-oriented

(you also get this cool stem previewing UI)

[hard beacon]

@verbal sun : you can find the answer in the FAQ channel 🙂 short answer: no

Quick question... is the stem export charged at 50 credits immediately or only when I want to download it?

[lunar mesa]

we charge when generating both batches of stems finishes

(you get two batches to pick between each time you extract)

[hard beacon]

ok.. I'm going to test this now and I hope that the function makes my brain explode

but first I have to find the best version of the 30 outputs 🙂

[lunar mesa]

metal!

[hard beacon]

oh.. now i am curious

[lunar mesa]

wait we need you to sign a waiver first

just kidding, good luck

😄

[hard beacon]

😄

What just occurred to me... are the stems in mp3 or did you change them from wav to mp3 because of the size?

[lunar mesa]

can change to wav from this menu

[hard beacon]

ok.. question answered 😄

[lunar mesa]

(it's a big download)

[hard beacon]

ok.. will search the best output version now to create a stem download.. will be back soon.. 😉

[lunar mesa]

let us know how it goes!

[hard beacon]

i will.. trust me 😉

[rough thunder]

so when ever im generating a song with a persona i made in 4.5 even if i change the lyrics and the description of the sound im after. it still keeps producing very similar songs music style vice that is

anyone run in to the same problem and have a solution ?

[hard beacon]

oi... i got an 0:09 output and no refund ??? tz tz tz .. system broken 😄

[polar spade]

Do the waves forms eventually show or is that just my using my Android display? Also, did I miss where the outputs reside -- or these tracks are only for download?I used a song with vocaloid which is notorious for overbleed on instrumental. No issues at all.

[brisk citrus]

Hi, it's tuesday and the answer is still absent. How can i contact the community manager? Let me know, thanks

[crude jetty]

most of my generations where I'm over 80% on weirdness is sending 10-20 second generations

[lunar mesa]

untick "hide stems" from the filters on the create page and they'll show up 🙂

[polar spade]

Oh on workpage.... I was looking in Library

I had clicked Download. Does that really mean download to library? Or download to my phone. I now see them in Library.

[lunar mesa]

high weirdness + complex/esoteric prompts can lead the model to make extra weird decisions like ending songs early 😛

[crude jetty]

A grunge beat with gothic elements, electric and raw 80/40 26 seconds

[lunar mesa]

"pop, edm" + high weirdness = pretty weird
"experimental alien laserfunk IDM" + high weirdness = ȓ̵͘u̶͌͛h̸͆̀ ̵̙͆r̷͗͛ò̸̕h̴͠

[hard beacon]

hold on.. did you deactive the remaster v4 version function ??

[lunar mesa]

it should download to your device, but takes a bit of time to prepare the zip

[unique stone]

I like to put weirdness to 95-100% to get some interesting "textures", and then use "Cover" at a lower weirdness to use that as a foundation.

[silver turtle]

🎵 "and you know what they say might hurt you..."🎵 lol

[polar spade]

I had to refresh a couple times before showing i Library. Just a matter of waiting a minute i guess. Makes sense.

[hard beacon]

stop stop stop... nooooo way... remaster v4 was deactivated right now.. no good decision..

[quick wedge]

You can send MikeB a DM and inform him of the issue that you ran in to and he’ll help you out from there.

[hard beacon]

that was my workflow to create good outputs from v4.5 personas...

forced v4.5 remastered versions ???? no no no noooooooooooooo... or how tiktok would say: hellll naaaaaaaaaah

[shadow plaza]

which option i should choose when i want to keep the original vocal and melody from my uploaded song ? Extend or Cover ?

[polar spade]

I guess I'll have to reserve judgements until I see if the only Remaster lives up to its intent. But yea, 4.0 Remaster has saved a few of these that 4.5 didn't work well with.

Else I'll feel like John Fogerty and having to recut every one of my old songs into 4.5 versions. Lol.

[hard beacon]

https://tenor.com/view/no-gif-20924249

[ebon sand]

mine doesn't, it's muffled but i never intended to use it anyway

[crude jetty]

some of my songs don't remaster well into 4.5, so that's gonna be a nightmare..

[ebon sand]

probably would save it as persona and generate it with v4 again to get a 'remaster'

[crude jetty]

there's no way to delete personas, or songs with one (even if it's the only one)

[ebon sand]

is that a problem for you?

[crude jetty]

exciuse the ping but @lunar mesa is it known issue you can't remove persona bases or the personas themselves?

[polar spade]

I'd say Cover as the simplest means if all goes well on the first try.

That's a 2 part question, assuming you are also trying to clone that uploaded voice exactly, which it may not and often doesnt do....

But it will give a nice singing voice and music and follow the melody exactly, at least when I cover an upload now for 4.5

[crude jetty]

sorta, because I had no intention of keeping the persona I created, I was testing something, and now I'm stuck with an unclean trash

[lunar mesa]

does the move to trash button work for you?

[crude jetty]

there's no way to completely remove it, just put it into the trash

[runic swan]

When I make a new section and add a prompt It never generates any music, no new edits after I click generate..... what am I doing wrong??? is anyone else having this issue?

[half dragon]

can't find a way to just close the new editor... other than navigate to a different page..

[round wigeon]

I am trying to sign up with my edu mail for this but I can only sign up with phone number

[hard beacon]

sign up or login?

[crude jetty]

I have not used the editor at all, so I'm not the one to ask about it, sorry

[round wigeon]

Sign up

[hard beacon]

well.. use your phone number, a google account or another email 🙂

[sturdy dragon]

Version 4.5 Pro, designed for professional users who demand the highest quality, creativity and reliability, costs 5 times more credits and takes 5 times more time to generate. It is exclusively available with the premier plan. This version utilizes the existing 4.5 model but with the highest quality settings—such as increased steps, no distillation, reduced quantization, etc.—ensuring the best possible quality even when the servers are overloaded. Because it leverages an existing model, implementation should be relatively straightforward.

[round wigeon]

Then this https://students.suno.com/#creation

How

I tried Google

with the edu mail

but nothing

[hard beacon]

ahh... suno for students is not available at the moment

[round wigeon]

Why?

[hard beacon]

https://discordapp.com/channels/1069381916492562582/1379186526864412672

[lunar mesa]

seems like the generation failed, did you get an error message after you clicked the generate button?

[runic swan]

none. ive tried like 20 times. and zero credits used

[lunar mesa]

ah ok i think this is a known issue

[runic swan]

okay

[lunar mesa]

we're investigating

could you try doing a hard refresh (command+shift+r on mac, ctrl+shift+r on windows)

[runic swan]

ill try

[ebon sand]

the old editor also doesn't generate if it matters

[runic swan]

that worked

[lunar mesa]

beautiful

will have a fix out for this as soon as we can, it might happen to you again

sorry about that!

[hard beacon]

@lunar mesa : lets check it out 😉

[runic swan]

Thank you so much for your help

[final rose]

#1: 4.0, fine audio ("mixing?") quality

#2: 4.5 sounds flat

This is unrelated to their latest update but I was hoping it would fix it. This seems to be all tracks I generate in 4.5.

And yes I tried exporting as wav.

[somber glen]

Uploading audio isn't working for me on Android. It'll let me upload some songs from Suno, some audio from other places but it's very inconsistent. Even recording small audio bits from my phone isn't recognized.

[daring lava]

Hey everyone 👋

As you all test out these new features, don't forget to check out suno.com/feedback if you want to! Best way to spotlight issues and share ways to improve! Thanks!

[versed oxide]

I cannot upload an audio file. There is no option for it. I am on iOS. When I click the upload arrow, I simply get a blank screen with a recording button and an indicator that it can be eight minutes. Does anyone have any insight?

[unique stone]

@bronze pecan an example of what Vocal+Instrumental stem split gives. This doesn't have much vocal in it to begin with but there's a choir at the start for example. Seems like it's using the new splitter.

Instrumental - https://suno.com/s/UTMl7yNpUlHShX3c
Vocal - https://suno.com/s/ZBBFMXUWePYnPoPl

Have a listen for example at:
00:15 - choir
03:07 - synthesised voice.

(posting here to avoid clogging suggestions channel)

This seems to give 2 pairs of stems costing 10 credits total

[hard beacon]

@lunar mesa Using the Stem function, I discovered the problem that the second part of a song often sounds metallic. While mastering, I was wondering why there was suddenly silence at the guitar stem. Instead of a guitar track, there was only a synth track. So, one component seems to have stopped working at some point during the output creation process.

the guitar track stopped at 1:20.

[unique stone]

It can sometimes put things in the wrong stem, depending on what it thinks the audio is.

[bronze pecan]

vocals sounds much better than what I got from my songs before. Hopefully at metal music the same 😄

[eternal pumice]

New song editor... I understand nothing... How to extend my song? How can I listen both parts and then join them?

[lunar mesa]

to extend:

then when it's done, click Save as New Song to render out whatever you have in the editor to a new track

[daring lava]

Lots or really cool articles just dropped in Suno's Knowledge Base too!
You can find those here: https://help.suno.com/en/categories/1453377-artist-update or #1294418290348921007 🙂

[eternal pumice]

And when i try to open my song in editor I am getting another song...

[lunar mesa]

you can also zoom in and drag the separation boundary around to dial in the transition

[lunar mesa]

that's weird, mind DMing me a link to the song you're trying to edit?

[clear crest]

This is fantastic. Now we just need the midi and then we are golden.

The editor looks amazing

[hard beacon]

@lunar mesa here is a better view.. guitar stops at 1:20..

[lunar mesa]

was this happening in both of the versions we generated for that song?

[hard beacon]

yes

[lunar mesa]

we'll look into it, thanks for reporting!

[hard beacon]

you are welcome.. if you want i can send you the link to the ouput via DM

[raven turtle]

I’ve gone back to 4 for this reason. I’m getting more dynamic results and the personas aren’t broken!

[lunar mesa]

a link to the original song would be great

[hard beacon]

sent

[eternal pumice]

Thanks for your help! I just needed to wait a bit for the extended segment to generate.

Now i understand how it works

[steep lintel]

I would love to have a remaster that didn't mess with the tone/EQ, compression, dB levels, or anything like that.

The degradation problems appear to be a little better, but none of my generations are actually usable yet and it almost feels like the only thing the machine learning has actually learned is how to mock me. The clipping in particular is probably the worst I've seen in my six months of being dreadfully disappointed with this remaster feature.

[clear crest]

I think the servers broke 😂

[graceful condor]

Wait holy crap SUNO is like an actual song editor now

That’s crazy

[plucky leaf]

I keep getting errors when I try to extend int he new editor.

[somber glen]

Can anyone tell me why uploads are not working on Android? I can't record anything and the uploads are inconsistent with what I can use and cannot.

[copper oasis]

I don't know if you guys have this too, but when I used the option to add sound and I have that sound, when it generates songs, none of the options creates a song similar to the one I added

[lunar mesa]

@plucky leaf might need to do this!

[oak siren]

same

extend feature is on the fritz for me. both beta and legacy

[lunar mesa]

oh - never mind, what are the errors?

[bronze pecan]

do you look also in Suggestions for browser channel for feedback on the updates? Looks like I posted my things in wrong place? 😐

[daring lava]

Best way to share feedback is here: suno.com/feedback 🙂

[lunar mesa]

sorry - just saw your recent posts there. understood and appreciate the feedback!

[past kernel]

The new editor somehow just gives me igibberish

[buoyant hare]

Hi guys ! I have a problem : I generated a song that I really love, the result is fantastic, but the first lyrics are missing 😦 Is there a way to correct that with adding the missing part ? https://suno.com/s/TomrcmA3kba0bTF3

[azure crane]

So you can't choose which stems you want to create with the All detected Stems feature? That means a lot of wasted credits if you only want to have one instrument for an example.

[buoyant hare]

Is the answer for me ? I don't understand everything ^.^'

[polar spade]

Ok. New update. You can upload up to 8 minutes.

[spice birch]

I've been doing that for over a week

They need to update the audio. I know it's still in Beta, but it shouldn't take much to take out that raspy Wasp poor sounding at the beginning of every song on V4.5. Most of the audio is turning out good except for the beginning of each song.

[tepid wedge]

Wow, if this new suno update works as promised (and keep in mind suno updates always underdeliver) then this is the biggest update ever

[desert shard]

New editor has option to insert part. I did not try it yet.

[desert shard]

Do not worry, it does not fully work 🙂

[buoyant hare]

OH ? 😮 How does it work ? I must get an eye in this tool !

What is a stem ? I don't get it

[plucky leaf]

Yeah, I can't extend in the editor or replace sections. I'm using firefox. Lets see if (sigh) Edge does any better.

[polar spade]

When you record a song in a studio, we record separate tracks: guitar, vocals, drums and so on.... we combine all those tracks to make a song. Those tracks at that point are called stems of the song. So Stem is used a a verb to also mean extracting out specific instruments and vocals of a song. The resulting tracks are called stems as well.

[buoyant hare]

Oh ok I get it now, ty !

I can't find the insert tool though ^.^'

[bronze pecan]

thank you ❤️

[buoyant hare]

ooooh ok I just found the access to the editor

[tepid wedge]

I knew it!

[plucky leaf]

Okay, it seems to be problems with the editor not liking to play nice-nice with firefox.

[tropic shell]

Hey all! Is there a way I can force the section I'm adding to be instrumental only?

[patent sand]

shouldn't remaster for v4 removal at least come after v4.5 is fully released? not that I used it (still stuck on free), but it seemed to fix a lot of people's issues

[desert shard]

New Editor has Insert and also Copy and Move. If they work as expected that is fantastic! New interface is confusing and not comfortable.

[buoyant hare]

I'm trying these, I'll tell you if I succeed in my mission haha

[torn spoke]

firefox is generally pretty bad at... most things

[azure crane]

Yes, the v4.5 remaster creates dull high-end that the V4.0 remaster could sometimes fix.

[patent sand]

other than that, the updates seem interesting

[lyric lotus]

Please for the love of God tell me ya'll didnt remove 4.5 to 4.0 remastes and this is a UI design issue?

[patent sand]

#🚀┃updates they did

exactly 2 hours ago

[lyric lotus]

I stg, that company actively goes out of its way to destory your generations sound quality, I can see it now over the weekend they probably had an all hands on deck zoom meeting "GUYS!!!! They figured out a way to get decent quality!! WHAT DO WE DO?!?" Shut that ** down, son"

[patent sand]

allegedly it was because of "low usage" of the feature

[lyric lotus]

BULL****

[daring lava]

Throwing this out there as you all begin to explore these new features...

Don't forget to check out suno.com/feedback if you want to! Best way to spotlight issues and share ways to improve! Thanks!

[lyric lotus]

They just launched 4.5 not long ago, and people were still figuring out that trick.. How can they say that?

[half dragon]

I'm sure they have the usage data to make the decision

[patent sand]

sending this message before they bring it back as a "demaster" 🙃

[patent sand]

can you help the guy in #archived-ticket-help ? I have no idea what an Audio Influence is

[lyric lotus]

I dont care what they call it, I just want it back. That's ridiculous

[patent sand]

I think the primary issue is as you said, it wasn't a very well-known feature. Most people won't think to remaster v4.5 to v4

[daring lava]

Thanks for flagging!

[azure crane]

That stem separation is really costly. 50 credits for All Detected Stems, even of u just need one instrument…

[vivid cedar]

new editor replace isn't working. generates for a while and then just stops.

[patent sand]

How in-depth does it go? is it pulling, for example, violin and cello, or just strings?

[azure crane]

Have no idea but I don’t think so.. I guess it will be like strings, horns, drums, bass, percussion etc..

[half dragon]

Yeah, but from what I tested... they're clean

[lyric lotus]

The new stem generations are indeed fire

[patent sand]

this was my main concern seeing the new button

[lyric lotus]

Guys, please bring back 4.5 to 4... paaaallzzzzzzzzz

[patent sand]

old stem separator could barely tell vocal from instrument

[raven turtle]

Vocal stems are amazing. I’ve had far worse sent to me as a remix pack from studios to work from. This is really getting there. Drums are little woofy and soft but vocals… damn clean

[lyric lotus]

I know, even when you're pulling vocals from a song with distortion on the guitars and stuff it sounds amazingly clean, I was literally just bragging on them before i found out about the removal of remasters to 4

[buoyant hare]

umph... I'm not convinced by the tool... The generated sections are really hard to control and don't really fit in the previous and next sections... :/

[runic swan]

I have a question. Im in the middle of a long instrumental solo and just trying to make it longer via inserting a new section... but no mater what I do it INSITS on adding lyrics.. and they are sung like gibberish.. Like what the heck why must it add lyrics when I literally tell it not to. I've wasted so many credits.

[sonic acorn]

I tried the new editing tool, and noticed there is no way to remove when you replace a section so if it refuses to do anything good (Like 95% of the time) you can't remove it, so you have to use the new ones you created which are full of hallucinations. So don't sit and edit a big part and get the ending working and then edit the start, you might ruin your start.

And I notice that still 4.5 is obsessed to end at the final vocal line with no ending music.

[lunar mesa]

sorry about this, we'll make it easier to revert to the original audio! for now you can always undo with ctrl-z or cmd-z

[lunar mesa]

investigating!

[patent sand]

?

[lunar mesa]

OOC are you in the default "Smart" mode down here?

[urban hornet]

Is it now necessary to have an annual subscription if you want to use the pro features? I can't use it when I'm on the monthly pro

[lunar mesa]

nevermind, not relevant for create section

[urban hornet]

i mean premier cant use pro feature

[quick wedge]

It’s probably a visual bug. Try refreshing the page and see if that helps or try a different browser.

[patent sand]

I'm free user

just confused as to the double Pro sign

[runic swan]

Actually it was jsut grey and stuck in classic. So I reset to default song and then I could select from the drop down.

[quick wedge]

I’m hoping it’s a visual bug or maybe it’s for something else. That’s my guess though 🤔

[runic swan]

Oh but then it instantly defaults back to classic as soon as I generate a new section and says "using same mode as selected clip" even though its a new clip @lunar mesa

[lunar mesa]

👍 thanks

we can make that a bit less confusing, working on a change that will hopefully get your instrumentals working properly

[runic swan]

see when I make a new clip there is no drop down option

[bronze pecan]

I didnt use it either in this time period because currently I'm not at this step of work. not because I'm not interested

but in my case its more about remaster v4 to v4

[runic swan]

whats the 4, 4.5 drama? I dont understand it

[lunar mesa]

👍 that's correct, create section always runs in classic mode at the moment

[half dragon]

With enough requests... they'd probably bring v4 remaster back...

[patent sand]

depends on how much trouble it would be to have that feature active

surely low usage isn't the only reason it was removed

[thick jay]

Can Suno be used to clone my voice and then be used for text-to-speech for creating narration for videos?

[gray vault]

I have a pro plam, but when i try to download stems i am prompted to upgrade. even the upgrade page says my pro plan is active. help? suggestions?

[slate trail]

What are the formats to upload besides local recording M4A and MP3 scene to not work I'm using this from my phone

I literally cannot upload anything besides record things.

[half dragon]

No cloning voice on Suno.
You can upload your voice and Suno would then create music and vocal sound close to it.

[patent sand]

with which you notably cannot create a persona from

[gray vault]

do you think this feature will be adjusted? I have my own recordings I'd like to create a persona from for various reasons.

[runic swan]

Is it possible to change context window size?

[desert shard]

There is option for undo and also reset, but you can just exit editor as all those changes within editor are not applied until you apply them.

[clever bloom]

Has anyone messed with the sliders yet?

I'm curious how the "Style Influence" one works

[half dragon]

Haven't checked it out yet but I'm guessing...
It'd be like sliding up to stay close to your prompt (style) and down to give model more creative freedom?

[sonic acorn]

I did a bunch of stuff at the end, then thought I would wait to save it before I fix the beginining because I was not super happy with the begining, but then it would not go back. The Edit undos did not exist but the appliance was still there.

[desert shard]

Leave it. Wait for few hours or better days and try again.

[sour laurel]

i am on a student pro plan, but i cannot extract stems and basically certain options that were previously avaliable are not avaliable for me at all, anyone??

[lunar mesa]

Yep! + and - buttons under where you toggle on the context window

[gray vault]

Same. Not a student plan, but a pro plan. Can't download stems of any kind. Sent an email to support and billing and have not heard back. I'm guessing it is a bug, but frustrating and definitely stopped my workflow

[bronze pecan]

wav works

[slate trail]

Thank you!

[clever bloom]

Hopefully that's the case

[alpine perch]

I have a question why do we have a word filter when it doesn't work

[mild scaffold]

i haven't been able to get the editor to work in the last few hours, has something happened? do i need to log out and back in?

Is there a browser I should be using?

[blissful rampart]

chrome

[jovial dune]

I could use a little help on something. With the new sliders, I was hoping I'd have better results than before, but I'm still not getting what I want.

I have an existing track where I absolutely LOVE the music but really want to redo the spoken words on top of it - some of the speech is mispronounced, poorly paced, etc., and there are a couple of dry spots where I'd like to have more speech. But anything I do to rework the speech dramatically changes the music underneath it, and often changes the whole structure of the song, which I don't want.

Song in question: https://suno.com/s/tHkZNKbofT0VV01d

I've tried both Reuse (with persona) and Cover, replacing sections, etc.. I tried extracting stems pre-update, but couldn't find a way to overlay new speech on the existing music track. And the new sliders are helping a little, but I still can't just match the music from the original track no matter what I do.

I could just take the music stem and manually edit in my own vocals (I might end up just doing that if there's no other way), but my goal was to keep this on Suno.

[blissful rampart]

are you a pro user? I am yet to tinker with the new song editor, but it looks as if you can highlight lyrics and simply edit them

[jovial dune]

In trying to match the original song as much as possible, my latest attempt sets Weirdness = 10%, Style influence = 70%, audio influence = 95%.

[stiff acorn]

Why would they remove the v4 remaster?!?!?
The only reason to usage may have been low is because we couldn’t set the model in the app. V4 remaster is the only way I’ve been able to find to get a good sound on v4.5 songs.

[blissful rampart]

@jovial dune can you not replace the lyrics in the new editor which should leave the music underneath in tact?

[jovial dune]

I wasn't aware of that. Playing with it now.

[lunar mesa]

drag to select the lyrics you want edited, then click "New Lyrics" and type in the lyrics you want to add instead!

[mild scaffold]

This editor has not been working for me today, when i hit replace it pretends to generate and then does nothing

[jovial dune]

Well, so far, the preview for the replaced lyrics makes it sound like it completely mangled them.

[regal jay]

Just wantted to see how many people are complaining about the "new features" lol

"Wild and unexpected results"

[lunar mesa]

you might need to do a hard refresh - we're aware of an issue that sometimes prevents generating. press ctrl+shift+r on windows or cmd+shift+r on mac!

[blissful rampart]

i am yet to play with them. I like the new stems idea. Only issue is it creates MP3 versions. Would love WAV for better mastering/ I have noticed the site is slowing down. I appreciate the changes/updates, but might be worth improving the servers.

[mild scaffold]

i full on downloaded a different browser lol

[lunar mesa]

depending on the song & specific context, you might need to select a slightly wider or narrower range in the timeline

[mild scaffold]

And then tried going back to the legacy editor and thats also just not generating

[regal jay]

Would not mind if the UI was a bit more intuitive and had a better way to manage the thousands of songs people have instead of all the news features, but good effort nonetheless

[jovial dune]

I get that, but I mean that instead of recognizable speech, I got complete garbage.

[lunar mesa]

mind dming me some examples?

[jovial dune]

Like, it didn't sound anything remotely like the English I wrote.

[jovial dune]

Sure, I'll DM.

[lunar mesa]

ty!

[neon vortex]

So...If anyone wonders what happens if you drag the "Weirdness" slider all the way to the right? Well... I did it so nobody else has to.

Results? Well, they didn't lie when they said "Wild and unexpected"

Because this https://suno.com/s/2DBL1bPSVtXZnIpg

Was very unexpected. And wild.

[regal jay]

@lunar mesa When is desktop app and API access coming?

[livid trout]

1. the update says you can change melodies, but i don't really see something in the editor that lets you do that
2. i'm trying to rearrange a song but there's no abandom changes option ? Worse even, when I try to reload the song in the editor, it doesn't load it intact, but with the things I cut in an earlier session

[blissful rampart]

lol @neon vortex its like a reverse washing machine on a rollercoaster

[lunar mesa]

what would you want out of a desktop app? [re: api](#1373511928005005454 message)

[neon vortex]

Yeah I don't...know what in the world is going on. Im deffo downloading the .wav and slowing it down just for lulz - and yes, I do name my gens v1, v2 and so on otherwise I get totally lost on which is which when I can't decide on a specific gen to continue with

[regal jay]

More efficient UX. Native apps generally have a better UI with more features. Easier to navigate.

[blissful rampart]

oh i do to. The endless list of iterations is hard to navigate. So i make playlists and when i like one, add it to a work in progress playlist. Then add V1, V2 etc

[neon vortex]

I turned the Weirdness slider down to 70% and now it's just producing an instrumental track with no vocals.

[blissful rampart]

it needs more forest and icecaps. 🤣

[regal jay]

Richer technical documentation about all these wild n weird features would be nice too but I imagine it is a trade secret

[neon vortex]

Well, click on my link and you have proper technical documentation - but it's the fast version

[unique stone]

After a very high weirdness, try using "remix" with the speed adjustment, and then remix the new clip with Cover and a lower weirdness but higher audio strength. Can get some interesting base "textures" that way to build off of.

[neon vortex]

Audio Strength? Que? I don't see a slider for that. Is that in the remix-thingy-majinga?

[unique stone]

You should see it when doing a Cover (or Extend)
Makes it "listen" to more of the clip you're covering.

[neon vortex]

Ah, there it is

[desert shard]

Problem with 4.5 is that with change of lyric it often significantly change music too. Tonight I spent abot 50 regenerations trying to change one word because even if it managet to change word it also changed music not to fit the song.

[neon vortex]

Im trying to get like, proper marching drum beats, you know like, almost like drumrolls. Like a marching band kinda sound but im failing horribly. Any tips?

[desert shard]

with this new option of creating 12 stems it may be possible to create stems to isolate vocal and then alter just vocal and merge them back hoping timing stays intact (for Repalce it usualy does)

[urban hornet]

This is error? or i need upgrade to premier?

i have pro plan but i cant use crop feature, i must upgrade it?

[remote flume]

Is uploading feature disable atm?

All my songs are gray out from uploading atm? They're mp3 files.

Like this

[desert shard]

Find audio clip with suitable drumming, upload it and extend.

[rose thorn]

Is this the right channel for asking general curiosity/info about the service plans and updates?

[urban hornet]

anyone?

[gray vault]

same issue trying to get stems. prompted to upgrade but i also have a pro account. support emails have not been answered. 1 yesterday and 2 today.

[urban hornet]

so now we can only wait?

[gray vault]

i dont have any other solutions. i'll keep checking this thread periodically. One can assume all the issues people are having at the moment are all related to the new roll out of features.

[gray vault]

I've tried opening suno in Chrome, Safari, on an iPad, and an iPhone & all have the same issue.

[urban hornet]

It seems like it's just my bad luck, because with the other account I can use the pro feature, maybe I just wait till they answered my email

[unique stone]

Regarding support e-mails: It's best to wait for the team to reply before sending more e-mails as each e-mail resets your position in the "queue". There is quite a big backlog at the moment so it may take a while to hear back in any case.

[gray vault]

Good to know, thank you. the reply email i got right away asked for screen shots if i had them, which i didn't yesterday, but i took today and sent along. we shall see. hopefully it is resolved in the next day or so, i have recordings scheduled this week & i'm trying to get samples & demos

[lunar mesa]

@urban hornet @gray vault mind posting or DMing me links to your suno accounts?

[remote flume]

Are you guys having issues with audio upload as well? Or is this just me issue atm?

[gray vault]

uploaded last night without issue, but i was disappointed to learn i couldn't create personas based on uploads.

[urban hornet]

https://suno.com/@refonicproduction

[remote flume]

I was fine last night too, but after the update, all my mp3 files are blank like this

[west shale]

Can one have 'global' edits? That is can I save a sound sample in Song Editor and then use that section in a different song with the Song Editor? right now it seems My Edits only has samples for the same song in the Song Editor

[west shale]

I mean taking a section from another song and inserting in another song with the Editor, is that possible right now?

[magic shale]

yeah, this is possible now by drag & drop I guess... did not try but saw this too

try it

[west shale]

but these are different editors

different songs

within the same song being edited, yes drag & drop does work

basically I want to take a part of a song and put it in another (let's say an Intro for example)

[magic shale]

ahh ! is this possible to mix 2 songs???? that would be great

[west shale]

it would!

[magic shale]

I suggested this idea some months ago !

[opal orbit]

Is anyone else having issues with clips for extend and replace just not generating? It will say generating and load for a minute, and then nothing new pops up in the edit list. Am I missing something?

With this new update I mean. I've never had this issue before.

[half dragon]

Would refresh or reload work?

[golden plover]

How long does it take to extract the 12 stems? I am extracting one of the tracks and it’s not responding

Is this a broken feature?

[opal orbit]

I tried, but no dice. Is everything overloaded with all the people trying the new update maybe?

[blissful rampart]

I've noticed this feature and then download (even just stereo mix) take quite awhile to generate/become available.

[golden plover]

I tried tracks which used audio or tracks which were generated out of scratch and it is stuck on Extract (50 Credits)

Is it safe to close the “extract stems” dialog box?

[half dragon]

I think it should be safe, you can access that panel again in the Editor - the generations are rendering on server side.

[soft pawn]

Sounds like shimmer is back in a big way? Maybe I just got unlucky.

[remote flume]

I guess I'm the only one uploading my own mp3 in? 😔

[finite grail]

Anyone replacing a section but nothing changed?

Okay, now it shows the alternatives. Didn't show up the like 5 times I've tried

[finite grail]

Just had that issue. Suddenly started showing the new generated sections lol

[shrewd willow]

Can you do personas from the mp3 you upload?

[patent sand]

only if there is no vocal

if there is vocal, you need to do a workaround involving extend or remaster or something

[gritty badger]

I don't understand why Suno allows to upload some of my wav files and some not? Anyone know why?

[remote flume]

I can't even get my mp3 to upload anymore

[gritty badger]

Ok I think there is a bug with Chrome UX on Android. I just tried on my Mac's Chrome and it allowed me to upload the same file, which it didn't allow on my phone. I think this issue has to be addressed

[magic shale]

convert them to mp3 320 kbit! dont upload wavs !

[gritty badger]

Why?

[magic shale]

1. cause too big 2. cause no big difference. 3 cause often they are to big for Suno too accept

[gritty badger]

But that wasn't my point. It allows to upload on Mac but not on Android web. Weird...

[livid wind]

This update is awesome

[solid rose]

I have a question. Where can I find "match lyrics to audio" in the new editor?

[finite grail]

opposite for me. Since the first time I've used Suno a year back, I was always unable to upload from PC but phone works lol

[gritty badger]

That's even more confusing 😅

[half dragon]

maybe this?

[remote flume]

I have mp3 files. They are definitely not big, and I was uploading fine before this update.

[merry holly]

The student pro plan lost all editing features!

[sleek stump]

From what I understand, there are no current Student plans....?

[quick silo]

So, where do we do the "fix lyrics" thing in the new editor? This isn't making a lot of sense to me. It was easy in the old editor to fix lyric alignment, but now I edit the displayed lyrics and they don't carry over to the new editor. They also appear differently in the old editor than the new editor or displayed lyrics. General confusion all-around.

[sleek stump]

OK, now that I have had a chance to look at the new editor... 50 credits just to get the stems for a single track? Really?

[tender eagle]

it's probably based on compute, I dunno how they extract them, unless they're not extracting, but generating them, which would be a separate generation of multiple segments which is more costly,

compute be costly

[sleek stump]

I use stem separators, they dont require a lot of compute.

[tender eagle]

my guess is it is not "separating" stems, but generating each segment as a standalone,

which would be a lot cleaner, but explain the cost of compute

but I mean if you can do it, just do it yourself then eh? no credits there

[unique stone]

This doesn't work the same way as other stem separators.
It generates the stems, so there's significantly less bleed and dropouts.

[tender eagle]

that's what I'd do

yeah that's what I thought

[polar spade]

@solid rose. you can click Legacy in the new edit UI and align lyrics as well did before. Whether this helps, I dont know, as you then must go back to a menu on the output to get back into the new Edit UI again if needing to use those tools.

[tender eagle]

but hey can I get a little love the the webdevs, when you guys are generating the video can you add a little "download vtt/srt/whatever CC" so i can use captioning that you ALREADY have pretty much done. usually. so it's free on compute side, and textfiles are tiny little things.,

[unique stone]

I've been using UVR previously and when Suno can figure out the stem correctly (sometimes it struggles with certain kinds of instrument) I get better results from Suno now.

The old Vocal+Instrumental separator got upgraded too (still costs 10 credits, uses the new separator and gives 2 pairs of stems now so you can choose which sounds better.)

[tender eagle]

3 things suno needs -> side by side players. that move in time on the same cue. so we can actually listen to the sections as they differ (solo vocals / instrument / chorus / bassy / etc) instead of clicking a thousand times to try an line it up, to see which remix / stem set is best :\

[sleek stump]

I would love to compare the new separator to my local separators I use, but dang the cost is high. Sticker shock is having a pretty hard effect on me 😦

[tender eagle]

50 credits is high?

[sleek stump]

for a singl;e track.. ya

[tender eagle]

it's not a single track, its multiple tracks

i mean it is literally generating n tracks

[sleek stump]

no its a single song... / track as most of us refer to a song.

[tender eagle]

yes but it is generating multiple generations of tracks.

it's like if you asked an ai image generator to generate 3 RGB and 1 Alpha layer image set. , it's still 4 images, even if they look the same when ya mix them together

basically image stems

but that's why it is costly i wager.

oh yeah 3 things... now 4...

1. Mixer to let us listen to multiple tracks (remix etc) or any really. side by side with same cue, so we can compare remixes and whatever else at same positions switching back and forth. or whatnot. or even any arbitrary tracks from our lib.. Clearly this exists somewhat present in the stem extractor, loading up a multi set of songs alone, and allowing for mono track selection whilst playing

2. When generating a video, export a VTT or SRT format captions textfile and let us download them, the video generation already uses a caption timer to move the lyrics on the video

3. checkbox selection of multiple tracks to move them around, the ability to make messes is easy, but cleanup is really tough in the library and workspaces.

4. i forget now... i'll rememebr later.

i wonder how smart it was to try to use stem separator on a very nasty sounding dnb track

[dim mantle]

How do i change the lead singer in a song? I have the persona but i want to just replace the singer but i dont want to change the song or lyrics, how do you do it?

[sleek stump]

Cover and designate a different vocalist type was the way it used to be done

Ive changed male>female and vice versa with Cover

[tender eagle]

i haven't used the editor extensively, i probably should, usually just to clip and cull simlish sometimes but damn it's a lot more robust now

Seriously though, can I get some "yes we'll do that" on generating srt/vtt captions either or during the video generation, or just a csv or timings for segments lol doing it by hand it atrocious for translating to english in captions for some tracks ... :\

oh yeah, and i'll never stop asking, but when can we get a santized dataset prompt set example, like 2 or 3 genres, lyric prompt sets and style prompt training inputs? 🙂 hell I'll sign an NDA and make them myselves for you for free... I really think this is a major roadblock to fully accessing the prompting... 😦

[dim mantle]

ty made the song sound amazing

[merry holly]

Yes there is student pro plans, I am in one.

[green orbit]

I find it really difficult not to be able to create a song in the key that I actually sing in. I mean, it would be nice if I could actually sing on my own songs, but I can't even see what I want cause it doesn't seem to recognize that. Am I doing something wrong or does that just not work in sumo?

Does anybody take their songs and have somebody write up charts from them?

[modest juniper]

I have an original song that i tried to upload and it said the lyrics are copyrighted. That is hogwash. I wrote the lyrics, did all the music and sang the damn thing. I got proof. I have even used it a bunch of times through the last couple years on here.

[ornate spire]

after you extract the stems, can you modify just one stem? Or is extracting the stems strictly meant for exporting for manual editing?

[proud temple]

I was just getting used to the new UI, which had been a slap in the face for my workflow when composing with Suno. The new update from today is a punch in the face...

I'm trying to get cover to work as it worked before the 3 sliders got introduced. I hoped the default values would reflect exactly what cover was doing before this update... But I can't really tell, it sure feels different, but I wonder if it is.

The fact Remaster v4 is gone is also sad, I was often comparing remaster v4 with remaster v4.5, in many cases I obtained a better result with remaster v4... now I guess I lost that option.

[ornate spire]

I'm so confused here. It looks like there should be functionality to see more rows of soundwaves (stems?) that we can modify individually, but we can only modify this singular row? Why is it laid out like it looks like you can mix various stems?

[proud temple]

probably designed by an LLM... There's many parts in the UI/UX that looks like it's meant to expose some advanced features, but doesn't.

wait for a week and you'll get a new UI

[ornate spire]

It's just odd like, it's cool that we can split the stems - that legitimately should allow us to have a higher quality version with each instrument mixed separately but... there's just nothing we can actually do with the stems outside of downloading them and mixing them ourselves?

[half dragon]

Yep that's it... for now.

[ornate spire]

I would even be happy with an option to "rebuild the song" with the improved mixing from separated instruments

[proud temple]

and then be limited with what we can do with our own mixed version (but I think we are less restricted now. which is good if that's accurate. haven't tested yet)

[ornate spire]

is there any kind of AI mixing site that can handle multiple instrument streams?

[half dragon]

Not that I know of, but quick google shows:

[proud temple]

I admit for mixing multiple tracks I use a software like Audacity (and I ask ChatGPT for instructions )
That works in some cases, but Audacity can be a bit... you need to get used to it

[ornate spire]

yeah I have audacity, I just don't have time

need a shiny button to do it all for me 😄

[proud temple]

the promise of AI

[ornate spire]

what I find hilarious is some songs it literally separates a stem of just shimmer... I would love to just remove that stem from the song

[proud temple]

I guess... if you have the stem for shimmer... surely you can cancel it out. That's how noise cancellation works after all

[ornate spire]

I mean we can just download the stems and combine them without the shimmer stem. I just mean a quick option to do so within suno and keep the fixed track in my library

[worldly rivet]

Hello, do you know if it is possible to remix known songs in suno? I tried but unfortunately without success😭🙏

[half dragon]

To remix, you'd need to upload and it's against TOS to upload copyright music. So... no.

[calm apex]

Where in this discord would I be able to get help, for matching lyrics to an audio upload? The problem is vocals often start before a chorus or verse and I dont know how to tell suno to stop using vocals to lead in like that

[valid hamlet]

https://youtu.be/oSX70sVlsUg?feature=shared

[tender eagle]

what is going on with the replace section dialogue? it just loads loads loads, and poof gone... along with my 10 credits

[dusky snow]

Super frustrating that I can't cover my own songs because they're copyrighted. Would be nice to be able to get verified as a particular artist so one's own copyrighted songs would be allowed in uploads.

[wicked canyon]

Hey Suno-Team 👋

Ich bin im Pro-Plan, aber wenn ich versuche eine Persona zu erstellen, erscheint „Subscribe to create a Persona“.
Laut eurer offiziellen Planbeschreibung ist das Persona-Feature im Pro-Plan enthalten – siehe Screenshot:

📛 Mein Suno-Username: @wicked canyon
🧑‍🎤 Mein Profilname: Patrick Martin Poliartist

Könnt ihr bitte prüfen, warum die Funktion für mich gesperrt ist?
Danke euch 🙏

[magic shale]

you can do that with Riffusion

[magic shale]

das passiert wenn das Lied aus einem Upload mit Stimme entstanden ist

[velvet jungle]

Anyone know if when trying to make a sound effect does it go in brackets or bullet stars [chuch bells] or Church bells anyone out there know? Havent had luck with neither, yet people arenable to get it to work but of course delete there promts from the lyrics! I used (*) before and after church bells

[cursive shadow]

I would just get a bells sound effect from an audio library and edit it in to the song in a daw.

[velvet jungle]

Hehe 5hwtsna smart move there .... well Inwas looking for the ewsy way out lol... however its a smart move thanks 🤟

[worldly rivet]

Hello, I have another question: what exactly can we do commercially with the productions and vocal tracks created using Suno?
Are we actually the rightful owners of the music we generate with this tool?
Some record labels are reluctant to sign tracks produced in this way.
The commercial terms don’t seem clear enough on this point. Thank you in advance for any clarification.

[cursive shadow]

It's not really that hard mate, takes like a minute in audacity.

[velvet jungle]

Gotcha... Definitely will make use of that now that it makes more sense....mate 🤟

[polar spade]

So far, both seem to work generally for sound fx, but only if Suno is familiar with the effect. But "bells" tend to get you more of a handbell or xylophone sound.
[Church bells]
Or
[bells] [church] (bong)(bong)(bong)
Or
[bells] [church] (bong)(bong)(bong)

Any of these would be considered by Suno i suppose, but then it likes to give it back as musical and more percussive ringing bells, xylophone, music box, and so on.

[proud temple]

figured I should at least provide examples of what I meant with remaster v4 providing better results in some cases.

original v4.5 -> https://suno.com/song/cb5060c0-3846-4866-9f20-6df8de09adc9
remaster v4 -> https://suno.com/song/96360c6a-9eac-4eb1-bb19-fe1f8ce1ae7e
remaster v4.5 -> https://suno.com/song/3bea38af-3c6a-45f6-a8f6-4b61943f803c

in this case, remaster v4 was giving a cleaner result.

obviously, every generation is different, but I noticed at least for some styles, I could get more consistent results with remaster v4. Cover is different. Remix is different. Personna is different... They all have their usecases, and when one option disappear, the usecase disappears. (I mean, the need is still there, but we lost the tool)

sorry for the spam of embeds

[polar spade]

Suno owns the created piece, at which point if you create under a currently paid subscription, the TOS grants you a commercial license by deeming you the owner with all options that legally provides. In other words, you get what you get, not exactly get what you'd have if you created and produced it all yourself in your home studio.

In turn you grant permission for suno to also train and make use of the creation, and as well, anything else you might upload.

So... "rightful owners" is vague. But you can be called the owner by way of a license. Keeping in mind you dont have copyrights on the musical aspects , in most present cases, unless you find a means of filing for copyright protection.

I dont know of record labels signing any AI tracks. They sign artists. Maybe you mean Music Publishers and Libraries? Of which 100% will not contract for anything AI generated, including vocals. However, you can certainly pitch songs to artists, who can then re-record the song.

If the TOS is not clear, it is best to have it further explained by an entertainment attorney is your best option. You may be right and an attorney can point out those discrepancies and loopholes, if any. Anyone not seeking legal advice would be taking whatever risks are involved in not doing so.

[polar spade]

You're preaching to the choir - we're just users and peers here. You should direct your concerns in the suggestion channels or even toss an email to support.

They state that as they notices few were using Remaster 4.0 that that influenced their decision. Not sure that's a valid argument. If Im use 4.5 Remaster 10 times , but 5 of those times is alone, the other 5, I use in conjunction with 4.0 Remaster to clean up the issues that 4.5 was making, then I've using Remaster 4.0 33% and 4.5 66%. This looks as if I'm using 4.5 twice as much when its better to say I'm using 4.0 in conjunction with 4.5 at least 66% of the time.

Anyway, I, at minimum, simply like the option of using , let's say , a milder 4.0 Remaster that can calm down the hyper glitches 4.5 would make at times.

[proud temple]

definitely... if one v4 remaster gives a better result than 50 remaster attempts with v4.5... sure, technically v4.5 was used 50 times more... but yea... telemetry can be misleading sometime

[ornate spire]

"the old version was better" - said at every update of every single AI tool ever

[proud temple]

which doesn't mean it's wrong :)

[ornate spire]

not necessarily.. though usually when you go back 2 updates you notice a clear reduction in quality

[proud temple]

I like v4.5, but I still have usecases with v4, and even v3.5

[ornate spire]

people said 3.5 was better than 4

[proud temple]

it's like saying: people said acoustic guitar was better than electric guitar.
I think both can exist... why not?

[remote flume]

Suno team, please fix android chrome music upload. I cannot upload any of my songs since this update.

Since we can finally use the full songs now.

[polar spade]

I'm still using 3.5 for matters of live performances. I'm using 4.0 remastered in conjunction with 4.5, which seems to balance out issues. So, there are benefits with each model that aren't transferred wholly to the next update.

[proud temple]

the fact they made the v4.5 a nice flashy magenta, and the v4 a dull gray... that doesn't help with the analytics. especially given the fact it makes the v4 songs look inferior in the lists of songs.

[ornate spire]

the only thing I hate is when they take away our ability to use old versions of things.. like why do they do that?

[polar spade]

Its been working fine on two uploads today for me. You may need to do some personal troubleshooting and testing. Also, make sure its an upload of perhaps a home recording, or at least record something on your phone then upload it to test.

[remote flume]

It was fine before the update last night for me

I even clear cache and all on my chrome, keep in mind, I am using mobile Chrome

Not Computer

[worldly rivet]

Thanks a lot for your response — it really helps clarify some key points. What I meant to say wasn’t to criticize the platform outright, but rather to express some confusion about the overall purpose of creating music that can’t really be fully used or taken further professionally — especially when it comes to labels.

I totally get the idea of making songs for fun, personal use, or small occasions (like birthday tracks, for example), and that’s definitely a valid use case. But if that’s the main scope, it does feel a bit limiting — particularly for those of us who are interested in more ambitious or professional applications of the music we create.

So my point is not to complain, but to better understand what can realistically be done with the tracks made using Suno — beyond the personal or experimental level. I’m just trying to get a clearer picture of the tool’s creative and professional boundaries.

[flint jewel]

can you run this on amd gpu ?

[polar spade]

Try on your computer. But sure, there may be a compatibility issues when you combine all features of your phone , model, browser and display format. The update today knock me from using my phone browser. All we can do is report it and see what tweaks we can do ourselves.

[remote flume]

Yes I can use it on my computer. The issue is my mobile have all my recording and songs VS my computer.

Ty for confirming that your mobile is also not working. That makes me feel better that it's not just me. 😂

[polar spade]

Its run through suno.com website. Most commercial consumer PCs would run it, I'd suspect.

[worldly rivet]

Do you know how to delete a playlist? Thank you 🙏

[warm coral]

It is

[torn spoke]

Uhh what?

Did they break persona's?

[magic shale]

I wanted to make an identical/similar cover... but this seems more difficult now with the sliders !
What is your experience with that ?

[torn spoke]

^ if I was able to make a persona I could give input

[north oasis]

now that i have the stems how do i add it to the online editor?

isnt the point i can add it to the online editor as well?

[magic shale]

Who knows what is the best setting for getting a similar/identical copy of the song made before ?
Theoretically I would say 0-50, 100, 100 .. but did not manage so far

[torn spoke]

I don't think so

[north oasis]

oh oh well seems like a design oversigt if you ask me but ill just throw it in fl studio

[torn spoke]

are persona's disabled for you too?

[graceful condor]

I'm having this issue too!

[torn spoke]

I guess I'll make the persona from the persona menu instead

BRUH

Why!?

It's a cropped version of MY OWN CLIP

wait, you can't make persona's from cropped clips anymore!?!?!?!?

[glossy pecan]

I wanted to know if it was possible to download three stems from the same song into one singular file?

[cerulean oyster]

Audio upload not working for me on mobile

[torn spoke]

.... I don't really understand what you mean. If you got stems in the same file, it'd be one track

each track is going to be seperate files

[glossy pecan]

oh ok ok

ty

[tender eagle]

are you using the correct model?

can someone expertly tell me why extend is a dice roll, i mean, cool if it sounds bad, but when it gives me 30 seconds of comepletely different lyrics... that's not what i paid for credits for, if I leave the full lyrics in it, it'll make a 2 minute song, and even restart the lyrics half way through. but if I make new lyrics, to extend into, it simply "reads" the part I cut out of the song, those should be refundable just like the 5 second clips, if suno can't fix their system,. It's a paid service not a beta.

and continuing a song into the part i am literally trying to cut out, when ther lyrics are not anywhere in my lyric prompt... that's on suno, shouldn't be on my credits..

[torn spoke]

Can confirm, you cannot make personas from songs that are fresh out of the editor. Absolutely horrendously stupid

:/

[tender eagle]

oh what?

[desert shard]

Theoretically that is possible. WAV is actually container and can have multiple tracks. I made those in some applications. But it is not widely supported by audio applications. I guess it is more practical to heave them as separate files.

[mystic charm]

hi! I was wondering if it is possible to get text and audio embeddings out of bark?

[proud mirage]

If i made a song during the basic plan, and then i subscribed to suno and extended/remixed the song, do i have the commercial rights to the new extended/remixed song?

[flint elm]

It's nice to see improvement when I remaster my 4.5 metal songs. To some songs it does wonders, others get worse. But at least the sound overall improves. When creating a new song with 4.5 (before remastering) unfortunately the old issues are still there.

[tacit token]

As a matter of fact I am new and intend to use the paid version of suno. Before that I need to make sure that suno works on my laptop. I have a Lenovo ideapad 300 (intel core I5, 2,3 GHz, 4 MB Ram, 2GB Graphic Card, 932 (-155) GB free space, 64 Bit). Can I generate music with this laptop (I want also to use my voice) ? Thank you in advance for your help.

[hollow hare]

I doubt you have 4 MB of RAM unless that is absolutely ancient XD

Fortunately for you, Suno processes the generation on their side, otherwise that will likely not be enough for Suno's model. That being said, as you can gather, the model is not available to the public.

To help get started, please check out the following. Do not be afraid to ask question but I suggest trying something out first with simple mode then go into custom mode for more control

/help

[flint haven]

How to use Suno

Hi there!
If you’re ready to dive into creating music, head over to Suno.com.

For more detailed information, check out these resources:

• Making Music
• Billing
• Rights and Ownership
• Accounts
• Troubleshooting
• Mobile App
• Suno Server

If you have any more questions, just ask in #general-suno, where you can always find our community, moderators, and guides to help you with most questions. 😊

[wanton dagger]

The most obvious omission from the new tools is a "remaster section" option. OR to fix needing that in the first place is match the mix of the entire song to the first 30 seconds of a song which would have to be done on the backend while generating. There is still terrible degradation of audio quality as the song goes longer.

[topaz tiger]

question, just woke up to the new changes, can you not take a v4.5 and remaster it back down to 4.0 anymore? am i missing that option?

[eager hull]

I made a cover, that lost the link to the original instrumental track. Now it doesn't show to the listeners that it's a cover but in the other creators profile it shows as a remix inspired. I kind of know, how, when it happened. Suno team probably knows too, though I can describe if needed. Just don't want to spread the recipe of a bug publicly. When I open the interface of a song, I see the source but can't touch the button, it's greyed out. I made significant changes, yes, but would be fair to credit the original author and honestly, some other remixes based on his instrumental wandered far more away from his track but are fully connected.

https://suno.com/s/HS8BdVNDT5cXOAIx

[ebon sand]

Yeah it's gone

[topaz tiger]

ya , well that sucks , I will be canceling my PRO until 5.0 is released with better audio quality. I was able to look past 4.5 audio issues as I could take a 4.5 track and remaster in 4.0 to fix some issues , now that is taking away, maybe I'll be back when things improve.

[sour dove]

I prefer the cover option instead of remaster. Never gotten any good results with the remastering option.

[topaz tiger]

The more I dwell on this the more I'm irritated about it. I was working on a Track last night and had a good version come out in v4.5 but it sounds soooo muddled, like a lot of v4.5 tracks. Normally I would take that and remaster it in 4.0 to bring up the quality, bighting up the sound, make a fuller profile, but to out right just take that away while they know that 4.5 audio has been criticized, well that just bad development practice. Unless they now allow with the other new editing tools to mess with audio equalization (that I'm not seeing), this is a deal breaker and a cancelation of a paying subscriber.

[cunning mulch]

I mean, I'm on an ancient lappy that also has 4GB of physical...

anyway, is there any way to retrieve the other set of stems in the new splitting mode? I thought I would get both with 'Download All'

[hearty bridge]

stuck in upload

never finished.

every audio

only supported aac flac opus

before the update it wasn't like this

[shell lagoon]

why does my advanced bar cut after the style? cant scroll it down, tried changing display distance but still the same

after refreshing the page the audio scale doesn't even partly show

[hot trout]

I woke up today excited to dive into the new features, and then I read the response of the community to this update, and now I am wondering... should I just wait a couple of days/a week or two before I waste any credits? Seems like there's a lot of jankiness going on for a lot of people, and while I love Suno, I still remember the transition from 3.5 to 4.0 and what a @#$! show that was, so I'm hesitant.

[dim mantle]

Every time we edit a song it takes credits. Every single cut and then Generate, replace, Generate, credits used. It seems wasteful to make people send money. Im sure thats the intention.

[magic shale]

@daring lava @steady kelp @lavish ice we need some instructions for using the sliders !

For example what is the best setting to make a cover of a internal suno song sound identical or at least very close to the original?
So far this was not a problem. No it is a problem !
(The last slider comes also in covers, but cover is not mentioned in the tool tip)

[hasty onyx]

Please! Bring back remaster with v4. Many of us don't like the sound quality of v4.5. Please, we're begging you!

[balmy junco]

Please, Bring back the previous suno or if its possible fix that REMIX force on every single song to be an option for turn off because when you work on one specific song and you want it with the specific vocals and apart from the instrumental it has nothing with it when its remixed, and now we don't have option for NOT to remix songs!

[dim mantle]

How to replace the intro because it keeps using the same one. Replace button keeps using the same riff. But when i Remix it remixes the entire song when i only wanted the one part changed. It uses the same song intro from another song. Its frustrating because its wasting away credits.

[torn spoke]

Anyone else having issues with distortion? I'm having an IMMENSE amunt of distortion in every section of a song where the energy increases

Done so many tests

The weird and style parameters aren't helping

:/

It's like the figurative speakers

and permanently blown out

[dim mantle]

I think that happened to me once but there were two tracks of the same playing.

If you have two, they go blairing loud

[hasty onyx]

For me, the problem is that version 4.5 has been out for a few weeks now, but the sound quality (it sounds flat, lifeless) just doesn't do it for me. While it brings some improvements, I still find the sound from version 4 more pleasant. And now that they've removed the option to remaster with version 4… what a letdown! To the point that I'm starting to consider other AIs. It's a shame.

[wanton dagger]

Sounds like you are talking about it clipping. Usually does that for the whole track not just parts, best bet is to just remaster it. Or if you like the rest of the track as it is but only want to remaster those parts, crop the clipping section then just remaster that part. Then put it all back together in a DAW. Only way to fix it as of now.

[supple parrot]

Hi, guys can you please help me? i'm pro user i'm trying to use my own voice as a persona, i uploaded an audio record but when i want to create the persona its grey and its a little messege like 'subscribe to create persona' 🙄😓

[torn spoke]

You can't create personas with uploaded voices

sucks but, its life

[supple parrot]

oh okay, thats sucks so i have to ai my voice and chose what most likely the closest?