#📚┃suno-school
1 messages · Page 65 of 1
aa
is my lyrics too cringey lolol
cap
cap
cappity-cap
i wear a cap and not a hat
i am not the cat in the hat
red and white, black tall cat
look around your shoulder
or i'll hit you with my bat!
supposed to be a meme song ngl
I generated a song, some unnecessary sounds were added at the end. How do I cut and end the song at some point at the end?
, black tall cat i'll hit you with my bat! ...... edit this, you'll be fine The cappity cap lines aren't helping, but by themselves, they'll pass.
external - audio editing software . Internal - extend with [outro] [end song]
Thank You.
oh ty
Suno doesnt let me use V3 when I try to extend a song, can someone help me?
Are you extending an uploaded audio clip or something else?
no its a song I made with suno
If I click on extend on one of my last songs in the create tab, i can change the version
but if i extend something from my library i cant change it
Not really for suno, but more for discord. Do u think Discord should change just one feature? Like when there is a ping in a category, should that rotate to the top of the left sidebar menu?
https://www.suno.wiki/faq/style-and-lyrics/duets/
If it helps, it's not a "you problem", duets are one of the most actively tricky styles to achieve -- there's a lot of separate "intertwining musical threads" that make a duet, so while it's not impossible to create through prompting, most times I've seen duets achieved it's through some degree of sheer luck, or a LOT of extensions put together
Just seconding the question that Furrybeans gave - if you're extending from a clip that even STARTED from an audio upload, you've lost the ability to use anything except for v3.5 on that track
That said, if you're unable to extend from a standard generation directly from Suno, it's best to get the ball rolling eary by emailing support@suno.com with the song link and the explicit steps you've tried -- what you did, vs what you expected to happen 🙏 💪
ah that is good to know, thankyou for the info and I will do my best :)
I
I've tried using [beat] to add some space between verses and such, but it doesn't seem to work. Anyone know anything that works? I've put in like [guitar solo] but it adds way to much. I just want a breather in there.
[Break] sometimes works for me, but it will typically cause a stop to some part of the melody or drop the drums, might not be what you're wanting though.
worth a try.
[-] works as does [interlude] as does [instruments]. Of course not always. Which means, sometimes, I literally instruct [interlude, right here, RIGHT F#CKING HERE, no singing, do some of your stuff on instruments, PLEASE, just do it.]
Of course, it leads to songs like this... https://soundcloud.com/sildona/low-down-mf
I'm pretty sure you can hear "interlude" in there along with the rest of my lyrics.
For what it's worth [guitar riff] seems to result in a shorter moment than [guitar solo].
oh yeah! I do that sometimes. Great tip though.
And smaller nudges are periods (.) At the end of a sentence, followed by a line space. One line space gives a little breather, but moreso, suno sees a blanking and considers what the next groupings should be rather than a continuation of the previous groupings of lines.
If not just an instrumental banger, then put in echos and comments or interjections such as (woo), ( breath), (growl) ,(gasp) to see if it's lightly thrown in but enough to give a slight pause.... and of course [pause] and [end] in the middle of a song just tells it to give a gap usually, rather than to actually pause or end the song. At least that's been for me.
I'm sure others work just as ell such as [scratch] and make up some like [rip] or any sound followed by FX like [gong fx] or [elephant fx], we'll, just anything to nudge it to give a gap.
Hello All, is it possible to make Suno just speak text fragment? Any tips for that? I am trying to make an intro where few phrases are just spoken not singed. Thx 🙏
To force a singer swap, use [key change] or [vocal key change] and variations of that.
absolute legend! thankyou for the info
Are you patient? If so, just keep trying. Use tags such as [Speaking] or [Narration] and variations to coax it into doing that. Also, ("parentheses with quotes") will often cause the singer to speak something. This is very much trial and error though. As with everything. The less specific and detailed you are the more likely you'll get a good result, in my experience.
Sure. Bear in mind this is just a reliable way to make it happen sometimes. It's not a guarantee. But I found most of the time when some random dude started singing it was because I'd tried to do a key change for the Melody, after a bridge or whatever. Using [Music Key Change] fixed that. And now the only time I get an accidental voice switch is when I use tags designed to enforce volume such as [belting], [Crescendo], or [explode].
If Suno is expecting a volume change or crescendo at those spots, those tags yield great reults. If not, it just changes to a new singer.
mhm alright, bit of a learning curve but loving every bit
This only works for masc to femme changes though. Haven't yet found a reliable way to force two different femme voices to alternate.
have you tried putting 2 female singers / female duet in the type of music?
adding thingslike that have been working well for me thus far
along with key change that could do it 🤔
It can be done but yeah, you're going to work for it. Try some of these styles and play with it.
[Removed links]
what are [key change] or [vocal key change] good for? And problbly can you give me example? I would like to try it xd, Thanks
I haven't wanted a duet.
Did you read the rest of what I posted?
alternating duet I mean
Same answer.
ah thought you meant that with "Haven't yet found a reliable way to force two different femme voices to alternate."
I did but I mean, can you use it like from scream vocals to quiet?
Just, you know, in my experiments. I never wanted a dude either. 😀
You can do whatever you want. Will it work? Only one way to find out. I'm just offering some notes based on my travels.
I don't have examples because it's never something I want so I delete them.
If you want 2 females in one song you can try name them like 1st [Anna] and 2nd [Sarah], I made duet like this, but with boy and girl. I went:
[Verse-1]
[Girl name]
[Verse-2]
[Boy name]
but it took me like hours to get them sing girl her part and boy his part
I did that the first time I tried with little sucsess, might have screwed up order though, good to know I am gonna revisit that
Tinda! 💜
will it work? only one way to find out
imo, so important to keep a sense of "wonder" while chasing these kinds of things -- all of the cool things that I enjoy about Suno are largely "mistakes I learned along the way"
Even if you don't find what you're setting out for, it genuinely saddens me seeing people play with a literal science fiction robot who wants to make music for them
, and they just immediately turn around and shit-talk the robot for not understanding something humans do, or not delivering "familiar but new" well enough.

has anyone any experiences with making "crowd clapping" as a Intro?
Yes. Don't, unless you want to spend a thousand credits. Edit it in afterward.
hmm
In theory it can work, but you're not likely to get a good song also, so it's kinda pointless to spend the time and credits trying.
agreed that would be the easiest way to handle it, especially adding something directly to the beginning of a song or the end, without it needing to "interplay".
Not a guarantee, and not at the beginning, but if it NEEDS to live on suno.com both Tinda and SixStringToker gave a shot at Suno standup comedy, so [crowd applause] could help
spoken word, comedy show, standup comedy, live audience, live performance song. Listen and make your own with Suno.
I want to make effect like crowd clapping as a intro u know
Alternately, generate a good intro, edit in some audience clapping, and upload that as your starter track. Extend it and see if you get something you like.
Yes. I understood the question.
Nothing stopping you from trying to make magic. Maybe you'll get lucky.
bcs I have [Narration] guy saying something like "...Let's clap together..." and then music start by "crowd clapping" and probably drums or something
I gotcha -- Tinda and I gave a couple suggestions or helpful tips to point you in the right direction, be sure to reach out here in #📚┃suno-school if you have any followup questions or want help from the rest of the community as well 🙂 💪
I mean, that's a good start. Other than doing it for you, I'm not sure I can help more. Just experiment. Have fun with it. Hope for a happy accident.
But remember I warned you that it would be expensive and probably not get you where you wanna go in the end.
Sometimes, though, the journey's the thing.
Persistence is at times its own reward.
ye ik it will probably cost ethernity but result will be sweet if it comes
One more recommendation if you insist on shooting the moon. Wait until a couple days before your subscription renews and blow all your remaining credits then.
Yes... but please bear in mind the end result will absolutely be better if you do it the other way (add crowd noise after). That's a Tinda Lee Guarantee.
I love my end-of-credit-cycle creative sprints, just so many different kinds of experiments, so many great songs to use as intros to extend from, weird ideas that normally I wouldnt want to "Waste" credits on, etc
ye I will probably make something like generate [Narration] solo, so I will not have any music there, next generate [crowd clapping] and then song
Legitimately something to look forward to each month
and I will acually try to learn how to use [key change] and [vocal key change], it might help me with that song
Bonjour ! I'm sorry if this has been discussed 2 M times, is there a place where you can read about the technology used, sample rate (is the sample rate higher on paid plans ?), and projects of sound quality improvements in the future, stuff like that ? Thanks !
guys how do you write [vocal keys change]? something like [vocal key: B Minor] or?
Hello
If I generate music in free plan
And then I can by pro plan
Can I Distibute Music to Music platform free track with our AP?
Can you tell me if you've been able to achieve a duet? I've lost almost 500 credits in trying.
The basic thing I want to achieve is a male singing a verse, and a female another verse.
tried like: [Verse 1 (Male)] , [Verse 1 MALE voice], or putting at the beginning: [Important: verse 1 is male voice, verse 2 is female voice.]
even putting "male female duet" in the prompt, it doesn't care.
now I don't know if the model is retarded or what.
try this
it worked for me
I used my 50 points already but if I find anything I will keep you updated
[Verse-1]
[Jack]
[Chorus]
[Jack & Anna]
[Verse 2]
[Anna]
it doesn't care.
always starts with the female voice
low-quality ai.
unless "jack" is a female name. i think the problem is definitely with the model.
Do not use the r-word here
better?
mby one day there will be option to make 2 voices that you will be able to add to song and give every voice what they are supposed to sing
.<
but yea thats true
one song where girl and boy sing took me atleast 4 days to make
yeah, it wiill take a lot of credits or a lot of luck
Is there a way to make Suno, when creating instrumentals, to don't make the lead line instruments SO prominent? They never stop soloing
usually using the word "interlude" makes a shorter section. Like [Flute Interlude] but, there is no magic bullet, every prompt you use still has a chance of not coming out the way you want
Zoth do u use key vocal change metatages? if yes how?
i've only used it once, but I couldnt tell if it changed the key or not
Does anyone know when the next major version update is planned for?
no
if anyone actually did know they would have had to sign an NDA so they wouldnt be able to tell us anyway
Yeah or it could be in a press release I was just curious
they haven't announced anything
do you write instrument like "turntable" into style of music and then also into lyrics like [Turntable solo]? does it have any effect?
general consensus is that when you are including an instrument that is not common in the Style you are using you should include that specific instrument in the Style Prompt as well as using it as a meta tag in the lyrics window
oh okey, I wrote it into style of music and also into music
I keep getting "couldn't generate that prompt flagged for moderation" its been serveral days any i still not seeing why it was denied? I am trying to make a parody song of the classic RamRanch but the wiki or non of the gudes say why can be said or not said. Should i try out the other competitors?
if it's full of vulgar terminology it won't allow it. If there is some vulgar terminology it will probably make it through but a song that is just RAMMED with it wont be made
i just tried searching for each offensive lyric on Suno and none of them are flagged and used with out issues but yet somehow it gets flagged which seems so bizarre. Its like i have to make songs that have to be kidsbop?
like i said, if the song is predominantly vulgar words then it wont be made, if it uses some words but they arent the entire song then it probably will. Ramranch is pretty much all vulgar terminology so if it's anything like that, then, it wont be made
so one of the songs I made is popping off
am I allowed to post it on youtube ect if I don't have a subsciption or am I only allowed to send it privatly to those I know ect? how does the copywrite work
and if yt posting is allowed, do I need to do anything specific revenew wise? though I doubt I'd earn more then a dollar but legally speaking
What’s your handle?
I might need a little refresher on what a handle is ^^,
just to add a bit onto this
I am not exactly planning anything big, just want to share my song in a few servers I'm in that only allow yt links
was just told that if I don't monotise the video it should all be good, correct me if that is wrong please :)
If your song was made without a paid subscription, you can post it on YouTube. You just cannot make money from it, and have to credit Suno.
so that means i would need a subscription and then i could upload on youtube or even spotify?
i just tested sumo on pc and would like to hear the music when im not at home too
No sub needed for putting on YouTube provided it's not monetized.
Anything you plan to monetize or put on Spotify has to be created during an active subscription
okay thank you for your quick answer and good night or day, depending where you are from:)
i don't know if this has been posted before or if it's common knowledge but i was very happy with the results when I generated an "instrumental" but wrote in the description "vocal chops" or like "male voice sampled through a telephone with static" and i was able to get that kinda EDM sound of a short sample that doesn't take up too much space
oh i just meant like on suno so i could listen to your stuff 🙂
You can go crazy in the lyrics with instrumentals, just use mostly unsingable things. Either [brackets] or non-words like punctuation or morse code.
yeah that's a good strategy too
i'm saying tho i was making "instrumentals" but putting in the description at the top stuff like "vocal chops with lots of effects"
i.e. not in the lyrics. both are good approaches
my problem with doing it in the lyrics is you have to "fight" more because it wants to sing all the time unless you tell it not to with [interlude] or something
so i usually have like 5 interludes lol
Hi everyone! Is there a way to set a specific BPM for a track in Suno?
Not specifically that i have seen. you can try listing the style as 'techno' or 'minimal techno'
What model is being used to generate the cover art for songs?
is it possible to generate only part2 + part3 from the song without part1 ?
Hopefully soon
thank you!
我的歌曲生成不了了不知道为什么,积分还有呢
My song is not generating anymore, I don't know why, and the points are still there.
您能检查一下自己是否登录了正确的帐户吗?也许可以清除缓存和 Cookie?
我正在使用 Google 翻译与您交流。对于任何沟通问题,我提前表示歉意。
Can you check to see if you are logged into the correct account? Maybe clear cache and cookies?
I am using Google Translate to communicate with you. I apologize in advance for any issues with communication.
trying to long in through Discord I'm able to sign in! but now this screen? (NVM it fixed itself)
So I needed to drop down to the $10 subscription but I wanted to keep ana ctive sub. ANyway apparently this caused me to loose 3220 credits because I went from 5700 (ish) to 2500. Aam I misunderstanding credits? I thought they didn't expire or get wiped?
The subscription page does clearly state renew, meaning that the credits do not carry over month to month, but resets/refreshes to the number of credits for the current plan you have upon the monthly renewal.
However, this situation definitely sounds like an edge case that may have not been considered.
Please email billing@suno.com with as much details as you can and they will fix the issue for you 🙂
Noted and thank you
Oh, another note: top-off credits does get carried over, unlikely the subscription credits, but can only be used when you have an active subscription (and only after you have no current subscription credits available for consumption)
so those credits if purchased will not expire then? Perhaps that is where I read the 'don't expire' part
Yep, the top-off will not expire, but they are much pricier than switching between plans
nodnods I can see where I got confused at least. I'll still email them tomorrow tho thank you for the great help!
SUNO doesn't know what MONO means. (Not that it knows much of anything, but just saying) This would be very useful since a LOT of the horrible glassy sound in most renderings is in the phasing stereo field. Removing it with phase isolators always introduces chirping artifacts depending on how narrow you make it. Still in its infancy but would be useful.
And in other news, it knows what a didgeridoo is, and sometimes how it sounds in normal circumstances. Quite interesting.
Another attribute that is so bizarre is the envelope (and attack) randomly changing so that a flute (sustained and breathy) becomes quite percussive and changes into a marimba sound and back in the middle of a melody. It's given me a whole new respect for marimba arpeggios and whatever weird toy piano Suno was trained on. Today, I got strings becoming percussive as well. I wonder if that mistake will ever change.
How to cut off beginning in generated track? Internally.
Hello I suggest, if you haven't tried it yet, an extraordinary AI. Personally, I input my raw song texts into it, and according to a specific model provided by Suno (Structure title, style, verse, chorus, etc...), I ask it to format them according to the model I provide Really, try it, you will be delighted (I hope)... here it is: https://chat.notdiamond.ai/
Not Diamond isn't like other chatbots you've used. Not Diamond automatically calls the best model for any prompt and improves in real-time based on your feedback, continuously learning your preferences. Not Diamond is the last chatbot you'll ever need.
From the Suno Blog, the remix contest:
We’ll be dropping some cool new features over the weeks, which you will also be able to use to put your own spin on Soundclash!```
The contest ends the 30th
I think we will be able to trim the part you want to extend, why else put a stem online for the remix that starts with 30 sec of silence https://suno.com/song/bd874965-820f-4a19-8506-48e37c53d049
There are lots of open contest questions, it's just the first day, things will get clarified. The announcement clearly hinted at relevant features. I personally suggested a nice tweak would be to add stem combinations (VOX + FX + DRUMS for example) to the contest playlist to give users more options even now.
Ooh, I'll check once I get to my pc
Though I do edit the songs further after suno has done it's thing so only the raw songs are here
Alrighty, thankyou :)
alright, name changed myself from whatever you start with XD
my handle is now Nilsbuss
hope you enjoy :)
can some help please. Im wanting to create my song from an uploaded 1 minute audio. i create part 2 etc. but how do i get it so the first 1 minute is excluded as i dont want the 1 minute instrumental at the beginning of the track that i uploaded.
Suno doesn't currently have that capability - if you combine the parts via "Get Whole Song", the upload will be prepended, up to the point that you extended from. You'll have to export the audio via wav or mp3 and use an external tool to trim that bit off.
thanks for that info appreciate it.
There have been a bunch of requests for that; maybe we'll see something soon(ish)
yeah hopefully. Thanks again for your quick reply.
i just add in the description 105 bpm .... or whatever.. seems to work for me
chip
I made a song on the free trial, it's not clear enough so I didn't use it. If I 'extend' it now with 'paid subscription', can I use it 'commercially' without infringing on your copyright? Because I love the beginning, beat, tone etc and I don't want to watch it waste.
We have discussed this numerous times here, no one knows the answer. You would have to ask support@suno.com
alright. I am new here. moreover i tried messaging them on instagram DM and comments and noone responded. Hopefully the support email will respond.
is suno not working for you guys? every time i hit create nothing happens
Did you run out of credits?
does anyone find if they generate a lot of song variations in rapid succession, it seems the quality of each successive one decreases? or is that in my head? Do you have to give the AI time to "think" and not "overload" it?
This shouldn't be a thing. As far as I am aware every time you generate it just creates a "job" for the server to handle. And it just has a bunch of jobs from everyone to process. If you take a break from generating for a period of time it'll just be busy processing other jobs.
Additionally the workload on the server side doesn't influence the quality.
I agree this would be bizarre but one of the (outdated) doc sites says you need to let it “cool down” or something to that effect
something odd happened, I noticed that I had 25/50 points left
how did I regain 5 points? XD
I'd be interested in knowing where you read this. It's never been a thing as far as I am aware.
By refreshing the page, that sounds like a visual glitch which does happen.
The most likely scenario that you will get a non-visual glitch for a +5 credit is if a track generation failed or was considered too short (< 9 sec) and you get the credit refunded (5 credit per failed track; with create making 2 tracks per submission)
ah that explained it, I did have a few odd creations XD
Curious if anyone has noticed commas being used more as suggestions to pause rather than a distinct precise pause, where the comma in the lyrics gets ignored more than followed?
One of the wikis lemme se my internet sucks rn
And any degradation might be just a fluke of suno giving too much balance to all prompts (especiallybif more than 5 prompts) on that generation. Thus it can't carve out anything specific from the initial static "noise field" and so you're left with, well, static and white noise- to put it in my words. -- Just a guess of course.
With just one prompt, it's like telling suno to just find the gold vein in the mine. Much easier to do. Just look for the shiny gold things in the rock. Ignore the rest.
One of the wikis lemme se my internet sucks rn
Hi everyone. I can't seem to be able to see the older songs I created with Suno. Do they get deleted after a while or what is going on?
Hello there. I have a problem. Every time I try to connect to my Suno account, the site creates a new one, and I can no longer connect to my account, the one where I created songs. I connect with Discord, I don't know if it's a problem, but I'll specify it. My account is the one here https://suno.com/@jexii
Make sure you're logging in with the correct account (might be stating the obvious but worth checking)
Hello, how do I get credit for suno, I would like to put music for non-commercial purposes on the following platforms
Ty for the answer. I am logged in with the correct account. It lists the newer songs I created but I can't scroll up or down further
Ah, are you on the Create page? If so go into Library - there's a page selector at the top right
This was it! Thank you. I panicked there for a few minutes lol. You saved myself a heartattack
Is it possible to get Suno to do a 12-bar blues? When I tell it to do a 12-bar blues (or just blues), it generates a "bluesy" song but it is not an actual 12-bar blues (the progression is more like contemporary country with a blues twang).
Suno implicitly has rights, titles, and interests in all songs made on their platform. 😲
If you make a song while you have an active paid membership, they waive their rights, titles, and interests in your song, and they're not involved in the ownership-side of things. 💪
If you extend from someone else's song, they're still considered an owner, regardless if you extended from their work with a paid account or not. 
In the instance of extending from a "free Suno song", Suno still owns the rights to the original that you made while the account was free.
💼
Email them with the specific song link and ask to acquire the rights to it. 📨 🙏
Ok I think I misunderstood a thing about adding audio. I made a song and I wanted to add my voice to it. So I got stems, and then I uploaded my voice, and whatever I tried, I never was able to add my voice to the stemed melody. I read the FAQ and watched videos but it seems I don't understand something.
Hi, I'm new here. Maybe someone asked here before, but I couldn't find it. So I'll ask. Maybe a month ago, when you uploaded the audio with the singing, it was not possible to make it public. But it didn't write the text or lyric or whatever it is. When I wanted it to use the same vocals (like CTRL + C and CTRL + V but with music) it did, but now it writes the lyrics and when I want to use the same vocals, it doesn't work and makes a mess. Does anyone know how to do it so that it is again with vocals like ctrl+c ctrl+v? if anyone doesn't understand, I can send pictures of what I mean (sorry for my English)
How do i get screaming vocals? I had ( gritty, aggressive, heavy metal, male vocal ) and I got awesome screaming vocals... I used the same tags and now all I get is Bing Crosby or Bing Crosby with a sore throat? Ive used ( scream, screaming vocals, screaming male vocals ) nothing is working....
no i have lik 700
when i sign on a different email it works but not the ones i have crdit on
But the question was who has the rights to the new song created with the extension. And Suno doesnt answer this question anywhere on their public pages
It's a combination of all of the authors.
You own the last bit, Suno still owns the free bit. It's collectively owned between you and Suno, the same as if you extended from one another Suno artist, it would be collectively owned by the both of you.
If you think "I own 100% of this song because I extended from artist_name with a paid account", they still have a degree of ownership & authorship over the clip,and would be in their rights to issue a takedown if you try to publish it without permission.
I dont think that's how legal ownership works, that you can own this piece of a song and someone else owns another piece of a song.
I'm not giving legal advice, I'm trying to answer a commonly misunderstood situation.
CONTACT LEGAL REPRESENTATIVES
but this is non-suno stuff, afaibt 👋
What suno owns is the file that was created on a free account, when you use an extension you create a new file.
So if someone takes one of your songs, and slap a 2 second extension on the end of it, does that bypass your ownership, authorship, and potential claims to copyright?
Or would you still feel like you had the right to protect your work
That's why it's not a black or white. Ask a copyright lawyer for your area if you have specific questions. Great to pose them here, but @flint haven has reminded me MANY times that it's not a technical issue and doesn't want us talking about it here
It's not black or white, the answer is: we dont know.
Suno doesn't explicitly say "who owns the song" because they don't get a say in that, they can only excuse themselves from rights, titles, and interests. It's not their responsibility to explain extraneous legal matters, and even go as far as to state that they don't guarantee that any output will vest in a copyright.
I wouldn't agree with "we don't know", because there's copyright lawyers who are paid (some of whom work pro bono 👀) to help explain these matters.
I can already feel @flint haven glaring at me, and I don't want to get onto his bad side, so I'm gonna step out, but I hope y'all are having an amazing day ❤️
and those copyright lawyers are not here in this discord
Discord is a social media platform, and shouldn't be a final-source for legal advice 🙏
Which is why we shouldnt be telling people that we know who has ownership of the extended song. Because we dont know.
...this gives me flashbacks.
Guys, is it happening to you that he repeats phrases from previous verses and goes back in the lyrics?
yeah, it happens when I don't have long enough lyric, or it happens to me when I have more bridges, or if the Suno says "i like this so i will repeat it". it happens to me most often before or after the chorus
But I have enough lyrics, I've been doing this type of musical genre for months and until now I haven't found this problem, it's a bummer because it's a point burner!!
and does it happend after extending the track? Or it does to you normally
@unique stone
https://sunoaiwiki.com/tips/2024-05-06-mastering-song-extensions-with-suno-ai/#_top
Use AI feedback to refine lyrics and composition before final adjustments:
Discard less successful attempts and allow the AI time to refresh.
Allowing the AI to refresh can lead to better outcomes when adjustments are made.
Explore how to master the use of extensions in Suno AI to maximize music production quality and minimize degradation.
AI needs time to chill too ya know they're really stressed out
Normally, 8 generations and only 1 turned out well, all the others repeat phrases after the first chorus or second verse, but just like that
Ah heh. Figured it might be that site. It's the source of a lot of stuff that I'm not convinced actually works, but people rely on it.
that sight actively displays things we know do not work, like negative prompts
"Discard less successful attempts and allow the AI time to refresh." - this is kinda meaningless. What would be refreshed? Why is it necessary? At the same time as one user is getting bad output, another user could be getting good output.
I know i heard it was a bad resource but he asked
Thanks - good to know that it's not a different source spreading this info, just the same place I'm familiar with.
I think that one was co-authored by ChatGPT Hallucinations.
Im getting an error saying i dont have enough credits despite having 9800. Any ideas? (Already refreshed page & signed out & back in)
it's sort of conceivable that there's a global seed that gets rotated or something, but that's really a stretch
So I think I misunderstood a thing about adding audio. I made a song and I wanted to add my voice to it. So I got stems, and then I uploaded my voice, and whatever I tried, I never was able to add my voice to the stemed melody. I read the FAQ and watched videos but it seems I don't understand something. any advice?
It just reads like the author was consulting ChatGPT on Suno features.
RE: the contest, why in all that is holy does the Soundclash drums have 12 seconds of dead air preceding the first drumbeat? (That initial cymbal at .01 does not count)
because that's how it is in the song
dumb question but I'm on 5 credits now - how the heck do I spend 5 credits lol so I get the 50 a day from the premium
you can't spend only 5 on song generation, so you'll just have to wait for the refresh
oh
Anyone have any success in having an overarching lyrical line over more rhythmic lyrics? Only example I can think of is The Police's Don't Stand So Close To Me where Sting sings "please don't stand so close to me" while Sting, Stuart, and Andy sing the chorus under it.
Hello is there a way to delete audios that we've uploaded, there's one that I posted by accident & I wish to remove it
You would probably be best served by emailing support@suno.com because that is something which would need to be handled by someone in the Team.
Oh ok I'll try that, thanks
Just for future reference to others... If you try to log in to the mobile IOS app with the google login connector and it asks you to make a new username it can desync your entire library and account subscription. The way to remedy this is by contacting support and weathering the storm of explaining yourself a couple times. Ultimately it was resolved and I am good to go! Putting this here in case someone else encounters this and breaks their paid account.
Is there a guarantee that the new feature mentioned in the contest will be released in time for us to make and post songs?
I think I misunderstood a thing about adding audio. I made a song and I wanted to add my voice to it. So I got stems, and then I uploaded my voice, and whatever I tried, I never was able to add my voice to the stemed melody. I read the FAQ and watched videos but it seems I don't understand something. any advice?
This isn't how the upload works. You upload a file, and then extend from it, suno will create additional music from that file that is inspired by the file.
I keep getting really weird sounding songs like they are just talking with a glitchy sounding voice
and it uses my creds
i end up having to delete 80% ish of the songs i make but no refund
not to be negative but hoping to fix it
GUYS! I'd love some feedback on my prompt skills — Would love to check out your prompt style as well!
https://suno.com/song/76ea144d-0390-4e2f-b4d4-5c82e5bb2da3
[Intro]
-----------
[RISER FX]
[Verse 1]
[Drums Entrance]
[HIGH PASS FILTER FX]
-----------
[Bridge]
-----------
[Chorus]
-----------
[RISER FX]
[Verse 2]
[HIGH PASS FILTER FX]
-----------
[Beat Drop]
[triplets]
[RISER FX]
[Chorus]
-----------
[RISER FX]
[Outro]
[REVERB FX]
-----------
Reggaeton, Latin Trap, Latin R&B, Urban, Rhythmic Pop, Latin Percussion song. Listen and make your own with Suno.
anyone have ideas???
I haven’t figured out a way to prompt for risers, sweeps, FX, etc. I think “echo” is the only one that works reliably for me. But you could try [buildup] or something and it might work.
I doubt if it understands “triplets” but maybe I’m wrong
Risers are a hit or miss. But in this track they were mostly applied to all sections in a subtle manner.
Havent tried with sweeps and other FX (besides HPF)
It doesn’t. Originally it was a hi hat drill with triplets (or something like that), but never got them to work.
Left it as “triplets” to see if the model would get creative with some kind of triplets but nope
I think it actually says “high pass” at 01:38 🙃
I love when it does that lol
Good tune tho
is suno having a stoke today? for the past 2 hours, every song I write or extend, its not following the lryics 100% like it was all week and last week OR, it is changing the entire song on extend.
this was not happening last week
The only filtering I noticed taking place was low pass
Seriously, Suno. I've tried dozen times to get duets. The one time I do not try I get six voices (maybe more, not sure). It's like Band Aid, just a bit smaller. 🤣 If the music was better, I would be thrilled. Probably never going to happen again.
ive tried a dozen times to get 3 voices and failed every time
Loool I didn’t notice
I'm counting ... I think, there are four and then a choir in the background. If you want to try your luck, the prompts in the style box contained: choir, harmony singing, male vocals. Feels like AI wanted to do something like Pentatonix. ^^
I'll try Harmony singing. I'm trying to make a song for the AI kpop band I am experimenting with Luna Flare. EP came out 2 days ago LOL
Definitely try "barbershop" to get some harmonies. It shouldn't sound like a barbershop if you tuck the prompt towards the end.
Interesting. Thanks. I'll give it a go. It's for a Christmas song so it could go with barbershop. 😁
Just noticed your name. Das von ... 😂 Good one.
Is Suno down right now?
I'm listening to a song just now, so it's working for me.
I have tried several times to extend a song, and keep getting an error message.
What does the message say?
Is there a guarantee that the new feature mentioned in the contest will be released in time for us to make and post songs?
Okay, that one never happened to me. Are you on your phone? I only use the desktop version. There I'd suggest to try another browser. Maybe the App has a problem. You could try close and reopen it (if you haven't done it already).
i had that happen a lot on the phone so I switched to desktop
for some reason suno is telling me i'm using inappropriate content but in my song there are no swear words?
ChatGPT is used to check the lyrics - it isn't just swear words it looks for but the context of the song. Usually the advice for this kind of thing is to paste the lyrics into ChatGPT and ask it what it doesn't like about the lyrics.
oh damn
well my song is vulgar to say the least
so i guess that's probably it
thanks for the help
See what ChatGPT objects to and change those bits 🙂
(Also avoid posting it in this Discord since we have a PG-13 rule here - just for songs shared, not for songs you make.)
yes i'm aware that's why i didn't want to post the lyrics initially
wouldn't be very smart
thanks so much 😄
how to achieve a sustained high note like this? https://www.youtube.com/watch?v=rF-DbZ6a4BE
Yo [finger snaps] works (sometimes)!
To my knowledge you cannot prompt for it. You can however hum into a microphone (even if you're a terrible singer) and upload it to davinci resolve. Now lengthen or copy/paste the clip to the desired length. Use the "Semi Tones" slider to animate the pitch. Now import it back into Suno and extend. Takes about 2-3 minutes, super easy.
For female voices slide the Semi Tons to the right. For male voices slide the Semi Tones to the left.
okay thank you
NP. And just a side note about that... thank you Suno Devs for allowing us to upload these files as videos into Suno. That makes life SO much easier than cooking them as MP3 or WAV.
Example (above comment about recording sounds)
Make any instrument, sound, effect, etc. with your voice and hten upload and prompt it. Magic happens.
Here is an example of me humming (making a neo-classical chanting choir). I sing like a screaming cat. I'm a terrible singer! But the AI fixed me and kept my pace and rhythm. What's mind blowing is how the AI took my hum and transferred it to the instruments. Just... wow.
Thank you. It kind of eventually worked but I had to really make sure to push the lute part. lute, lute, lute, lute, medieval lute minstrel music, fantasy, elven, tavern ambience, reeds, male vocals, ONLY LUTE I'll need to change the elven part to sylvan, thought I don't think that has any impact.
good morning everyone. how can i have only one voice that sings for me? i mean for everyone song that i create
Good morning. It's hit and miss. But you can try "strong female lead", "solo" or "professional singer". The words imply individual singer.
You can also type your prompt like this:
I'm spinnin the beats [male voice]
Feel the heat [male voice]
Droppin the rhythm [male voice]
Droppin the BEAT! [robotic male voice]
Does anybody know the email to contact Suno?
I may be wrong, and I hope somebody corrects me if I am, but I don't think there is a way to have the same voice over multiple songs.
I saw it on another platform. I thought it was possible here too
I could be wrong. I hope you can find out if it's possible.
For billing-related inquiries, contact billing@suno.com
Otherwise contact support@suno.com
Im trying to make songs and type descriptions in my own language but for some reason random words in my language are Names in some Artist.
Example
Само it means only. But suno detect it as Samo and it was Artist Name.
Same for кара it means force you. But same it was Artist name and dont allow me to use the word...
If this continue soon there will be no words left for me to describe a song 😂
Language is Bulgarian
Last night tried to make song about friend his surname is Draganov... You guess it Artist Copyright
Suno uses ChatGPT for both lyrics generations and moderation. Unfortunately, some people decided to pick some common phrases as their artist names (or producer tag) which is annoying.
To get around these issues, phonetically spell the word you want.
E.g. "live vs. live" can be differentiated by using 'liv' versus 'lyev', 'lie-v', etc. (Unsure how 'lyv'is stressed, can go both ways :X)
Similarly, the phonetic spelling for love would be luv, etc.
That may work for now. But it will be great if Suno dont translate litteraly the words from another language. I mean not even translation is more like 1 on 1 converting Cyrillic to Latin Alphabeticaly speaking.
Like Само and Samo.
Should translate Only
I hope this got update as right now. Rhymes in Bulgarian songs are out of sync. They dont rhymes at all 😂
I wonder if other Non english languages have same problem?
And the punctuation is wrong like 50/50 when they sing
Hm, ChatGPT really is not that great for lyrics, I do admit. People have mentioned other services for it but I cannot recall what they were (though searching in Discord should yield results).
Perhaps try rephrasing the request when generating, such as:
"Write a song as an award-winning lyricist in Bulgarian about ..."
Which should theoretically sound better compared to asking it
"Write a Bulgarian song about ..."
Etc.
Will give a try thanks
You can also tell it to do certain rhyming patterns too, like "... using an AABB rhyming pattern in verses..." (or ABAB, etc.)
Thank you!
I searched for 3 days and found nothing.
hi.how can i have a video like this video if i move level with music audio? https://www.youtube.com/watch?v=7obACOPmZHw
موسيقى عربية هادئة ومريحة بالعود والكمان والناي للتأمل واليوغا, موسيقى استرخاء وتركيز | قائمة تشغيل هادئة
استمتع بأروع موسيقى عربية هادئة تجمع بين أنغام العود والكمان والناي, مصممة خصيصاً للتأمل, اليوغا, الاسترخاء, والتركيز. هذه القائمة تخلق جواً من السكينة والراحة النفسية, لتساعدك على الاسترخاء وإيجاد السلام الداخلي. مثالية لجلسات التأمل أو الي...
The create button isn’t putting anything down on the site
I have credits so I’m confused
This occurs after I input or hit the create button
Hi
I wonder why all my remix for the soundclash competition cant extend more than 2 min? I have the v.3.5 on...
With v3.5 new songs can be up to 4 minutes and extensions can be up to 2 minutes. You can extend the new part again to make it longer if you need to.
Ok ty👍
Tho my problem is that the new part is also only 2 min
Each extension will be 2 minutes. But then you can use "Get Whole Song" to join them all together
Ok I will try
Tysm🙏
Cool! Glad it helped!
Sure there is... but you won't like it. (trial and error, grind it till you find it, etc).
I highly recommend finding a style prompt for your vocalist that you use every time. That won't make it automatic, of course, but I've been generating songs using the same "singer" for months. The key is, you have to be willing to toss out a great composition if the voice is wrong.
Hi, how do I know the date my subscription expired? Is it possible to see which songs were made during my subscription, so I can see which ones I can use commercially?
Each song tells you when it was generated. Bear in mind though, you need to look at the original generation if you used an extend or "Get Full Song" etc...
hi, isn't there a command for Suno to use the same uploaded sound without any modification?
Not exactly. Depending upon where you extend from, Suno may try to continue the original melody, possibly even including bits of the original audio... or (often when extending from silence) it may just use the audio as a reference and continue with something similar... or it may ignore everything and give you something completely new.
The last one doesn't happen too often, but it does occasionally happen.
and what about vocals? when you upload vocals, you can see the lyric, but one month ago, there was nothing written. Now it make mess, like changing melody of voice, or better problem, the lyric is wrong and making mess twice. If you know what i mean
So @naive quail has it right to say, the extended music generations produced by suno from the same audio won't be the same with each new generation from the original, if occasionally it does come CLOSE to the same style and sound.
i uploaded vocals a month ago it wasn't writing any lyrics and now it's writing and making an extreme mess. A month ago when I uploaded vocals it was like ctrl+c and ctrl+v, with music. Now it changes the melody of the voice, and that makes it out of sync with the music and it's all off.
".... When you upload vocals, you can see the lyric....."
It seems you say, if you upload your own audio, and that audio has singing in it, then the suno output will put lyrics of that vocal into the lyrics section? It should not be able to do that, so I'm sure I not understand clearly.
".... One month ago there was nothing written...."
At some point, you may have added lyrics to the generations? Suno will grab all lyrics in the chain of generations and dump into your final "whole" song, so you may have some generations blank, but as you add lyrics, all or combination of lyrics might show up. Maybe that's what's going on there.
I haven't uploaded just vocals in a while, but it should be the same deal. Suno treats uploads like any other suno track, so extending from them works the same way. Picking the right extension can make a pretty significant difference. If you extend from the middle of vocals, for example, you'll more than likely to get a continuation of those vocals.
Yea, I haven't seen that before, tho it's been a minute since I've uploaded just vocals. @worn thistle can you link an example or two?
can't upload becose it have uploaded vocals
i will send screenshoots
You're right. Vocals found in the upload will not let you post Publicly.
Do you mean because you're worried about sharing those vocals here, or that there's some technical reason? If for technical reasons, you can still share the link to the track, you just can't set it to "Public"
ahh right sorry i complete forgot on that opption
Gotta do a report for work, but I'll check back in. @polar spade, you got this!! 
You can...... and I and others often do.... is go back and edit out lyrics from the previous generation you are working on. Click edit lyrics, highlight, and delete all lyrics if you wish. That can help clean up things. But keep in mind: don't forget to keep a copy of your lyrics somewhere. I will even email me my own lyrics. Haha.
It can give you unwanted results or be exactly what you want. But it's a way to see if you can control suno from making a mess of the vocals, as you say.
If it makes things worse, then paste lyrics back in and try again.
Keep in mind, the more prompts you have, the more "production" in the uploaded audio (now suno has to pull out each instrument and vocal), the more lyrics you fill up a page (maximum is what, 3000 chracters) --- from all of those, the more suno will be strained to produce a clean, consistent output. Less, in this case, is better.
Edit lyrics? highlight? I mean, i need tutorial. Cuz i don't understand how.
Wait. So you uploaded a second time and suno automatically transcribed lyrics? This is not the same original output. So you 1) uploaded two times? And 2)the second time, word appeared in the lyrics box?
Suno does automatically transcribe the lyrics of uploads, even just speaking
it's kinda awesome
No problem. Yes, if you scroll below any lyric section you can click "edit lyrics". It will open the lyrics that you can edit, copy, move, delete.... then scroll down again and click SAVE. Don't forget to SAVE. I forget all the time and have to start over.
Yes, it wasn't there a month ago and now it is
See there. @worn thistle . That's something I had not discovered as I have uploaded music with vocals. @worn thistle so, there may have been an internal update that did NOT transcribe your uploaded vocals before, but does now. That may answer that.
So now, if that's true. Any outside vocals, we may see, show up in our lyrics. Definitely then, we would have to monitor our lyrics and make sure they are as we want BEFORE we do any generation. Suno will use EVERY thing found in the lyrics box.
yeah, i see it, probably. You mean "Edit Displayed Lyrics"?
Yes. Correct.
Hah. I didn't know that Suno had started doing that. Interesting.
One of your biggest challenges is probably going to be coming up with styles for the extension(s) that match the original
thx, now i can try, if it will work like before
I have not decided how much Suno "remembers", but by deleting lyrics, or even changing lyrics of previous generations, or the original...... it seems to help stop suno from reaching back and using random lyric choices from what it can find.
At least you have an extra tool to work with, editing lyrics, which you can see how it affects your work.
yeah, you know, I'm from czech, so if i use czech vocals, Suno will write in Poland. You don't know how frustrating it was
yeah, you know, i know that this tool existing, but i never find it, i always search thing is (vertical ... )
they put that button is a very non-intuitive place
@safe hazel thanks for the heads-up on the audio transcriptions. 🙏
no problem, it's super cool
it works It works It works, thx men,
i wanna say thank you very much
Dude. I'm just as shocked and chuffed that something worked. Lol. But you're welcome. The others @naive quail @safe hazel got it right for you. I just come in with the broom.
Kinda funny that they added that audio-to-text feature to add the lyrics to uploads... and the fix for extending that particular upload is to delete that added text. 
in first time, i was mad, because it was making mess. But now it's great again. I just wanna say thx to you too. Thx for yours patience guys
I would like to respectfully request the deletion of my suno account
We're all users just like you... we can't delete your account. You'll need to contact support@suno.com, or maybe billing@suno.com
Unfortunately they seem very busy
So instead I've come here to call them a bunch of slurs
How's that working out for you?
Very well so far
The longest that I've had to wait for a support email reply was maybe 4 or 5 days, but it's been a minute - the wait time may be longer now. 😦
just gotta figure out how hateful I have to be to get banned
Only way to delete the account
Nope. You'd just get banned from Discord and I would recommend not doing that. Your only way to delete your Suno account is by e-mailing them.
Done that, now I just have to wait 13-50 business years
Is this the right channel? So for anyone that is doing extensions on the Soundclash stems https://suno.com/playlist/ccf70d74-e47b-44f6-ab36-5ee5b9d39758 and made them public, can you find them if you later search for them on the "search tab" (not on your profile)
Extend one of these stems for the chance to win $25K and perform your remix at a live show with me, Flosstradamus, in NYC. More details at suno.com/blog/floss
Is there a guarantee that the new feature mentioned in the contest will be released in time for us to make and post songs?
new feature?
How about this then, we'll just time you out so you can't post "bunch of slurs" and you continue to wait patiently for your email to be responded to? Because as the Guide already said, the banning of your ability to be in this Discord has no impact at all on your account with Suno.com.
I think this is a brilliant idea actually....
Anyone know the ETA for the official API to be released? We've already integrated the placeholder API, and it works fine, but I'm worried about using it in production as it's linked to my own account
how many Nvidia A100 GPUs do you think or know Suno or Udio uses
No news about API that I'm aware of.
Using any unofficial API would be against the Terms of Service and may result in action being taken against your account.
Also it would eliminate those tracks from being allowed to monetize.
But that's a whole bucket of worms that one..leads to all sorts of problems.
Placeholder API?
Like Furrybeans said... I wouldn't use that if I were you.
There is no guarantee and no one here knows about release dates for anything.
Thanks, do you know if any suno dev sees this channel? Would love to talk with the one of them directly.
We are a VC backed company with a very popular AI game called Suck Up: https://www.youtube.com/results?search_query=suck+up
Please don't post things like that. Because they are illegal to use, and they never had any permission from Suno either.
Not remotely official.
Apologies, I was actually fooled by the description, for some reason thought it was from suno.ai devs. In any case, it's definitely not something I plan to use as it's clearly not official
Mods, can we get a frequently asked questions channel for tech chat? If i gotta answer one more question about how to get the same voice for every song im gonna go crazy
that would be handy
Not me. I just want to know how to get 6 backup singers in harmony. (He said, with tongue in cheek)
Or like pinned convos(?)
Would people read FAQs or pins though? 🤔
I mean, the website help/FAQs cover who owns the songs on free or paid plans and still the question comes up
but we could just refer them to the faq instead of having to explain everything again
yeah
plus I'd use a faq if it meant stuff with suno's stuff with song creation because that'd be handy
"hello I am new here and I have a question i am sure no one has asked before. Is it possible to get the same voice for every song on the Suno website? How do i do that?" Answer: "it's in the FAQ" much easier than having to explain it or ignore the person
There's the wiki but yeah it doesn't cover everything https://www.suno.wiki/
Also https://help.suno.com/en/categories/550017-making-music is useful
I like how I type comments in the vein of conversation and after reading, say.... nahhh.... and delete it before clicking send. Those solutions that you know will just open another can of worms.
I do this a lot.
It's a thin line between helping one person and just confusing 10 others.
I typed a short comment in reply to someone replying to someone else earlier and wished I hadn't because it seemed to cause more confusion xD
I'm just looking at the little boy actor who's clearly tired of doing so many takes.
Speaking of kid actors have you seen the kid in the last bit of BTTF3? WTF was going on there?
I can’t seem to finish a song. Although I signal the end, it keeps going for another 50 minutes or more. I need it to wrap up within 15 to 30 seconds. What command will stop it?
I also tried switching between versions, but it just keeps dragging on with pointless sections that don't finish the song.
Mods, if we had a FAQ section this would be in it
The wiki has a section with suggestions on how to get a song to end: https://www.suno.wiki/faq/making-music/how-do-i-end-the-song/
🤯 To the question: Can ChatGPT create a language from scratch, with great coherence and logic?
The answer is yes! I asked him, and it's a fact, he did it!
I'm still amazed!
Any known errors right now? My v3.5 songs are finishing at 2 mins (instead of 4) and now there's a system error when trying to make a song
is suno down right now? im getting an error when i try to generate somethinh
looks like it yup
:/
What the heck? My credits disappeared and it wants me to subscribe again? nvm, its back
Yep!
Can't extend my song at this moment
Looks fixed now
Yeah
Anyone else's songs still stuck at 2 mins on v3.5 ?
are you using an extension?
yeah but im only extending from 24 seconds
doesnt matter, extensions can only be two minutes long
The 50+ or so other songs I've made would disagree with you lol
they wouldn't but ok.
Can I screenshot in here?
yes
So how is it that I can have a 3:36 song that's a Part 2 extension, when you are saying they're capped at 2 mins?
it's because your other extensions were from an upload, that is handled differently
Well that now makes sense, thank you!
😄
So to fix one simply downloads and reuploads?
just keep extending
I'm trying to do it with vocals that's all
you can extend from extensions, you can also select "get whole song" and then extend from that
hi guys. how can I cut off the end part of a song
You can try extending from before that part?
noo man, for example i have 3-minute-lasting song and i want to cut off the 2;15-3;00 part of it. can i do it on free account?
or have to get subscription
or maybe i will download this song, and maybe you guys can recommend me some other program in which i can do this?
I know "extending" doesn't sound like a way to end a song, but trust me -- Hepcat is onto something.
If you create an extension @ 02:15, remove ALL lyrics, style prompts, and ONLY use "end", you might be surprised! Since you're not giving any extra instructions, Suno might generate a VERY short end of the song.
Related funny story, my aunt got a computer for her living room in the 90s, and then immediately hated it because "it was too loud" and she couldn't read over the mechanical clicking & whirring of the hard drives.
She called me one afternoon because she wanted to turn it off, but the gentleman at the store didn't tell her how. I told her to click "Start" and she bellowed at me "I don't want to start it, I want to stop the damned thing!" 😂
Suffice to say, "extending is the best method of ending" at the moment 🙂
but sirs i think you did not undersatnd me. i want to cut off the 2;15-3;00 part of a song and get the remained(00;00-2;14) piece of a song
is it possible?
The method I described will give you the original 0:00 - 2:14, plus a very short ending -- preferably an immediate fade to silence
ah ok Sir, thank you i will try it then !
You can very easily cut off the end of a song using an audio editor like Audacity https://www.audacityteam.org/download/
But the method I described will allow you to keep the song on suno.com to share with your friends this way 😄 💪 As you can only re-add songs to Suno that are 6-60s in length 🧠👈
Pop back into tech-chat if you have any questions, and there's bound to be someone here who has found success with this method 
Download Audacity for Windows, macOS and Linux
ok now all clear, thank you sir once again for this comprehensive answer. appreciate it!
What's a good way to prompt for a unique male voice?
Male voices tend to sound very similar almost everywhere
Female voices are definitely more varied
omg omg omg new feature???!!!
songs aren't ending for me they just cut off and I am wasting so many credits on good songs for that contest, that I cannot use because of the janky ending. Anyone else having this problem the past two days?
I cannot use the harbor method because we are supposed to be extending someone else's stem and then submitting that extension. If I take all instructions out that is not very creative.
You should keep extending, sometimes with or without a prompt. Use things like [fade], [end] in the empty prompt
Yes
So. I don't know. What's the cover beta do different from extending an uploaded audio?
It isn't currently possible to use the new cover feature with soundclash, right? Since it's not in your library.
i think it tries to maintain the same vocalist, at least the eaxmple they posted sounded like it
So far I had one cover completely disregard the original song and just make a new song with the prompt
and another one just repost the original song verbatim
lmao
Yeah... that happens.
Oh OK. I thought it did that already. I sang some la la la di dahs... weeks ago and thought it captured my quality, but I haven't used vocals in audio but once since then.
I'm not sure why... and sometimes you just get the original song back but slightly higher pitched or a different tempo
well you can use covers on your suno generated song as well
it's this
or maybe C as well
IDK!
Still just 60 seconds for an upload? I'll try one of my demos from years past. I guess it emulated the whole song but with altered production based on the genre? Rhetorical.... I'll just go try
I've always wanted to be able to put vocals on top of a preexisting melody without changing it
Ngl I don't see a lot of use for the covers feature, sounded cooler when I read it than when I actually try it (when it actually works). Maybe I'm just not seeing the potential though.
Covers is probably not gonna work for that
I tried a couple times and even with identical genre tags added lyrics completely changes the song
Hi all, I wonder if this is an issue that others have. If I try to upload a recording of playing guitar and extend it, approximately 80% the resulting tracks will have no drums on it, no matter how many tags for drums I add.
how long is your upload?
~50 seconds
cut it down a bit, the longer it is the more suno draws from it. You want the new song to be more unique by adding drums, something the upload doesnt have
try 30s, it will be more likely to add new elements
or 20s
you get the idea
Can the free users make their own cover art? The Ai ones look like trash or at least use Stable Diffusion models to make it better
I understand. I wish that to be something that Suno can adjust going forward though. I love the idea of being able to record some riffs and then use Suno to add the drums, bass, vocals.
Sounds like you want an AI powered mixing/editing application. I'm sure we'll have one soon.
Same here
Not sure Suno ever plans to go in that direction, but someone will

That would be a thing of beauty, generating individual tracks based on the context of what's in other tracks. Rerolling the drums/vocals independantly of the guitar for example. 
Or adding lyrics to a previously instrumental upload
Yep
That's a feature I would certainly use a lot
Yu, that was added recently. Previously uploading images was locked behind subscriptions, now it's just generating images with Suno's native image-gen
Having recently had a rough idea I recorded quickly transform into a really nice piece of music, I expect one day we will have a way to render individual parts of a song in a DAW into an AI-generated performance that fits with the rest of the song. Will be cool.
soon we'll just hook the AI up to our brains and it will make a song from thoughts
Soon we'll just upload our brains into the music and become the flow 
Anyhow that's enough random chatter from me in here.
My final thoughts on covers are: Kinda cool, not super useful in most circumstances - way too unreliable to be worth using (imo) in the circumstances where it is useful for now.
So far I covered a suno song four times and just got back the same song
Probably like 60-70% of my generations are either a verbatim repost of the original song or something that doesn't even sound remotely similar, hence this comment:
way too unreliable to be worth using (imo) in the circumstances where it is useful for now.
Did one and prompted rap, in the styles. Got 20 seconds of babbling rap, gunshots [or firecrackers] and then the cover song same as before.
I'll try a few more with vocals more out front. And fewer seconds. See what happens then.
sounds like a successful rollout 
I'll play with it , the cover feature, to try to find something niche and cool with it.
A useful thing I experienced with it: Prompting for specific parts like percussion or vocals, and not giving a style, can cause it to output individual parts. I got a clean percussion stem by doing this. Totally not the intended use of this feature but useful.
Also use the same style prompt as the original for a noisy/bad quality song and it can make it sound a bit better.
There's times where covers have given me "inspired" songs, where it seems to have ALMOST captured the melody, but ends up being entirely unrelated, but there are WAY more instances where it's blown me away by capturing fine minutiae that really compose the makeup of the original -- small trills that are adapted, specific timings and pauses, etc
Just like usual with Suno, there's a degree of learning along with getting the most out of the feature.
I expected a degree of "soullessness" from the covers, as I know how much effort goes into creating acoustic guitar singer/songwriter covers of Punk-Ska songs using traditional means -- there's a lot more than just playing the correct chord progressions, there's a "heart" to a song that absolutely gets lost when it's not covered "with love".
It's not 1:1, but I'm honestly blown away by the functionality of Covers, right out of the gate. The team f%@#ing nailed it with this one, and I haven't been so refreshed and excited about Suno since I first discovered it.
Gamechanger. 2 eras of Suno -- pre and post covers.
Idk. The only thing that really stands out to me is that the first 200 covers are free. Idk if that's per month or per account, but I digress... Idk if we could do this to others' songs but we could already remix our own songs by reusing the prompts and making edits.
The ability to edit lyrics without changing the melody of the song is what I'm really after. Say I have a song that I really like, but I've made a mistake in the lyrics, say a typo or soemthing similar(heh, example!). I want to be able to recapture the sound of the song without the randomization. Sorta like how one can use a seed in an image generator to get similar results.
I think that's what the covers are supposed to allow
https://suno.com/song/a019ac3f-3e5a-4a33-8b48-55a5ea137365
Polyrhythmic is in
I'm doing things that are insane atm
Holy sh--
When I cover my vocal stuff it just immeadiately deletes the tracks
(no style)-hyperpop #myspace, Harsh polyrhythmic trndyMouse, nostalgic, hidden sadness, broken composer song. Listen and make your own with Suno.
This is pure insanity
New music will come from this, it's like it's made by something alien
I love it
Good day I know it may be slim for the right people to see this. But individual work spaces for each song until I want to publish it would be nice. Also so I can delete all of the bad ones easily.
#💡┃suggestions-web #🚧┃suggestions-layout -- the devs read the suggestions channels for ideas! The tech-chat, not so much -- but I've been begging for workspaces, so I hope that you post your ideas in those channels where the devs can see that more people are wanting to see improved organization tools! 💪 🙏
I've just used the cover version for 3 songs. It made songs. 2 completely new songs with the same lyrics (which I can do myself actually with create when preserving the lyrics) and 4 similar songs with same voice and instrumental. The advantage in the 4 songs might be a new recorded, better sounding voice (with no degeneration). Is this correct ?
You can get a better quality version, a completely different song (which can be annoying because it's not what you expect at all), or you can re-style the song to a different genre, different instruments, etc. it can be hard to control though.
For example, here's a solo guitar piece "covered" by a koto: https://suno.com/song/8ecb44f9-292e-46fa-9f8a-d18ee445ac3d
the AI makes in cover all sounds new? voice and instrumental? If so, this can be used some day also to update a song to a later Suno version which might have higher sound quality
I expect that'll be possible. I've got some noisy old v3 songs that sound better now, and even managed to prompt to get vocal and drum parts from them after a few tries.
But it does take a few tries, because it will often give something very different, or the same song. Likely why it's beta at the moment 🙂
well, than here would be good to have also the feature to show somehow and also play in the same window the original track to compare cover and original
I'm not able to use the "cover" feature on vocal stems, as it doesn't appear in the 3-dot interaction menu -- can you explain what method you're using when you're trying to cover vocal tracks, and what it looks like when the tracks are added & deleted?
Audio uploads of my freestyle raps
i am a vocalist
so i was really excited because ive made dozens of tracks that cannot go public using my own material however they get instantly deleted upon creation.
however i was just refunded for these a minute ago
I'm going to test with an audio upload and see if i get similar results 🙏 thanks for bringing that up
edit - I was able to record a quick audio upload directly through the site and made some covers of it -- I would be curious the results you got were a temporary hiccup, or if you'd get the same errors later. 🤔
I'd be happy to use some of my cover credits giving it a try if you wanted some more data 🙂
I made a comparison, Udio vs Suno
I made a comparison bt. suno and udio.
If you are looking for as many song ideas as possible, the 30 usd plan and the 32 seconds gen from Udio will produce 4800 ideas for this money, the most, more than twice as Suno.
And with the 2 min version, Udio will produce 400 more song ideas than Suno.
To finish a song is another beer 🙂 This would require a lot of credits for the 32 second version.
Will Cover feature for free members?
Is it just me, or do Covers have vocals with a higher quality?
I wonder if it has something to do with the way they are made 🤔
I feel like they do.
So, does 'covers' mean I can finally fix up my older songs?
Looks like it. Cover + Reuse Prompt can increase the quality and remove "artifacts"
I'm listening to something now and one word had a weird audio artifact in the original but the cover lacks it
Haha and the first 200 covers being free of charge ✌️
I like that
That's 100 attempts
im not asking from nowhere they said it
how do you do both? the cover + reuse prompt... i dunno if i need more coffee but i dont see how to set both
the feature was literally released today
First select a song and press Cover, then Reuse Prompt which copies all lyrics and the style prompt, and finally click Create
In my case one of the covers always sounds almost exactly like the original + some enhancements in some cases
1/2 that is
i click cover and it takes it to the redo page i cant select both... I am on PC
one was. apparently there will be more. it was released after i asked. wtf is your problem dude
Reuse Prompt simply copies over all lyrics and the style prompt. Try inputting it manually
i think you should be asking yourself that question
ah ok i think i got it .. thank muchly
I'm making my lyrics tweaks
or trying to, anyway
why is it only coming in at 3:25 though?
No idea, most of my songs cut off between 3:00 and 3:26
the original song is 5:05
see, my use case is as a form of in-painting
I have some minor lyrics tweaks to add in
Does anyone know how Suno decides which songs to feature in Trending or New or Genre Playlists
Can somebody explain to me what's the difference between cover feature and basic extend option (with changing style promt)?
Put simply, Covers are a bit like audio2audio and they attempt to mimic the original song's melody and oftentimes pace
hmm... Cover is still using the vocal glitches
are the vocals exactly the same?
maybe the intent is to keep the vocals the same, while changing up the music
so it wouldnt fix vocal glitches
sometimes, it incorporates the change though
It did for me in a few cases
Yes it's a bit of a gamble
Like with all AI models now 😂
I'm glad we get to test without any credit costs
True, to get a feel for what to do
https://suno.com/song/36c4643a-ffd2-4503-8ec7-413a2ee33f79
https://suno.com/song/acbd81f2-9c9d-47cd-8945-0badb3be8bb0
This was my very first attempt after the announcement and I liked it
Oh the embeds, let me remove that
Ideally, I'd like it to have more weight to the new lyrics box
has anyone had some seriously bad distortion on new tracks since yesterday? two tracks I want to extend seem ruined by this strange distortion...
I was really glad for this.. and though i just have done 2 tries.. the second try was absolutely what i needed. I wrote a song a while ago.. did a bazillion versions.. fell in love with one.. extended it etc and realised i had included one of the verses twice. there was no going back.. now there is and with the cover + reuse ... i went in and fixed the lyrics.. I did have to extend twice because its a longish song.. but i am a happy camper.
it also appears on instru stems...
boop. what kind of distortion? I did get some weird stuff yesterday but i thought it was my weird lyrics
the boop is for TomR
So it seems like voice getting stuck, but instrumental has changed
its a steadily rising siren sound mixed with a shimmering crash sound that rises to an intolerable level and ruins everything... appeatred twice now on songs I badly want to extend
ugh .. can you try the new cover thing? that might fix it
what i m wondering ois has this been noticed or reported by anyone else?
Yes, a bit like audio2audio with the gist of the original melody being kept while making it different
and it picks and chooses what changes to incorporate
I think I know what you mean. It might be a model issue.
let me go llisten to my yesterday things.. i just thought it was me
Would Remix have been a more straightforward term instead of Cover?
Or did Suno want to be different from Udio perhaps?
sufi i dont know what that means... what mode;l?
suno used to have remix
Suno Models are the ones that generate songs. I do notice a sort of "distortion" with some songs, but not with real ones on Spotify. It's likely a model issue
So I don't really understand how the "Cover" exactly works.
But i still dont fully understand, i tried it a couple of times. it seems random. One of my trap generation track changed to some ballad
Well i guess i need to try it a lil bit more
IS it copying, how the lyrics worked on the past songs?
Yep I'm a February user, pre-v3 even (I tried Suno when v2 was around only) and it was called Remix before being renamed to Reuse Prompt
It takes the melody of the original like a form of audio2audio
Here's an example
Hmm, might need to try it out a bit more to fully understand it
So it's not taking the stem?
It's not taking a stem. It's more like creating a new song based on the original with changes you want
It tries to follow the melody of the original
Do I need to stay within the same genre bands?
Like, can I take a "pop vocals" and put it into techno?
I think you can experiment with any genre
even if you add exta chorus or prechorus or delete verses, it still does a good job.. and no you can change genres etc .. i think thats the point
Yes it's like a convoluted way to inpaint 🤣
I can tell it's Beta however
Take the original, cover it, change some lyrics
Sometimes I get 55 seconds long songs haha
exactly lol .. i love that.. inpainting music
yeah but for music lol
me too
audio2audio and inpainting are good equivalents in terms of terminology
Are you able to show me like a, graph of how it works?
I'm on mobile but basically you take any song, click "Cover", then change whatever, and hit Create. It can make very varied covers or be closer to the original depending on RNG
it seems to be limited to about 3:30
better .. just take a song you like and cover it in a different genre.. or take a song you want to change the lyrics on but keep the same melody and style
so it'll cram anything longer into that timeframe
i had to extend twice... i noticed that too
it cuts off..
and that's a problem, because extending... is extending in Suno
It still makes remastering v2 tests possible
it ended up fine for me.. but i didnt change the style in anyway.. i just wanted a redo because i put a verse in twice by mistake and forgot the chorus once.. and the prechorus three times.. i was able to fix and extend .. and am really happy now.. just fixing and mastering outside suno now
I even considered taking v2 songs and making full versions
They were literally just tests back then
well, I have a full, 5:05, v3.5 song I wanted to tweak lyrics in
i am considering it now lol
mine was 7 min.. because of my mistakes.. redone properly iut came to 5:54
https://suno.com/song/c09bfc95-0143-4d1a-8bbf-2e8cc74353f9
I tried it again with the same example song to clarify
I did choose genres that are closer to the original but the idea is that you can change things
And you orignal was what genre again?
house, dark, electro, female vocals, piano, edm, dance, emotional, sad, melancholic, clear vocals, bass
The cover I sent does sound different
Ah, so your original was edm related
emo, pop, destined, buzzing, melancholic, happy, uplifting, positive, happy, lo-fi, emboldened, arpeggiated melodies
So it seems like it doesn't like techno
trance, progressive techno, progressive club beat, banger, trap, high pace
This is not working out so well for me
Ahhh, so the difference between the two gens is Denise
maybe the original cadence doesnt fit well
Denoise
Let me try again with this style
So is the issue then, that my voice only will work with the pre-defined ones when it was made?
Yea, that's what I think as well
Which one is which
psst i am a sekrit yolkhead fan
One is a low power, one high
Cool pic btw ✌️
https://suno.com/song/e7e47910-6f8a-4762-a1a2-634578c18ed5
Is this closer to your preferred style?
Maybe it's all about the original song's tempo and melody
Yeah, it definetly is 100%
So that's good too know though
A darn shame though
Maybe it's RNG
Check my examples please:
https://suno.com/song/3cb4d288-a388-4cf3-95c4-52a16954e10b
https://suno.com/song/3426b524-5d4f-4928-97f4-d6dbc56c7907
Track just changed completely even when i tried to put almost the same style of music
Im doing something wrong lol
dark agressive trap, distorted 808, rage mood, screamo rap song. Listen and make your own with Suno.
Aggressive trap, ragecore, scream-rap song. Listen and make your own with Suno.
you can actually take your lyrics to chatgpt and ask it to redo the cadence to fit a different style.. thats what i use gpt for mostly .. i write my own songs then check the cadence with chatgpt
See, it works for you, since you still stick on trap
I was trying to go from emo, rock, alt-rock -> techno, and that didn't work at all
try redoing the lyrics cadence in chatgpt
Can you send the original song?
Ok, now it's working. I picked another song, and slightly shifted to a kind of related genre
Nah, the point is it doesn't work for me. Melody is completely changed
Went from:
soul, r&b, symphonic soul, rap, male voice, soulful, reverberation, lo-fi, guitar, soulful, ballad, reflective, romance
Over towards:
hip-hop, soul, rapping, moetown, 60's
And now it understood rapping, hip-hop
But this, was after 6 tokenizations though, it's beta after all, the other 5 before it, was just a copy/paste sound
Suno definitely dropped an interesting feature
I noticed that oftentimes Covers also add instrumental intros to some songs that didn't have them
So I wonder if I take a v2 song and add lyrics before the old ones, will it be like extending backwards?
Loving Covers for adding lyrics to instrumentals. It's working really well with changing genres, too. Great move, Suno!
Idk, doesn't it worked before with extend?
I did many songs this way. Instrumental with v3 and then structure and lyrics with v3.5
Must try this
This works though!
https://suno.com/song/2b0bc16a-4553-487d-bfc6-da2c05860028
Tried one of my oldest tests and Covered it
Added the "Intro" part as a quick test
I've never tried before!
Suno: You want inpainting, audio2audio, and extending backwards? Nah, you get Covers - which is all 3
Original song: https://suno.com/song/636a5d1e-85b4-427f-b8e5-d296949d771d
**Barbershop Quartet Cover: **https://suno.com/song/53b3d53e-9f6d-41df-8371-abd035846da0
soul, r&b, symphonic soul, rap, male voice, soulful, reverberation, lo-fi, guitar, soulful, ballad, reflective, romance song. Listen and make your own with Suno.
hip-hop, soul, rapping, Motown, 60's, old-school, barbershop quartet, A cappella song. Listen and make your own with Suno.
This worked out alright
So I guess, "stay in your lane" sort off
I selected cover so I could try to re-write some of the words, but it's exactly the same as the origional
The lyrics section does have the edited version, but the actual track is all the old lyrics?
Try generating again. The first 100 tries should be free
I tried like 10 times now
I added lyrics before an original song and it worked on the first try
I think this is why they gave us the free covers, since it's also in beta.
How does Suno recognize foreign languages? I.e. Greek
Ok, I'm starting to get it now, thanks for the help guys
it does Japanese really well.. so I am sure it would have no problem with Greek
Use google translate and it will work fine. I did an album in Korean and it crushed it
👋 💡 I'm not sure where to put this - but with the Covers feature it seems like we're training the model to create Covers.
What I'm noticing with my reject generations for "Covers" - is that it would make for a really useful "Redo" feature for my songs where I want to keep mostly the same melodies but not change the style.
I hope this is a good guess, but if I'm grasping at straws - I'd love to suggest that as a feature. Use the negative inputs from Covers to make it easier to Redo songs
google translate makes a lot of mistakes. deepL or chatgpt are way more accurate for translations
I dunno with Japanese, Korean, Italian i've had great success., we use it at my day job
Chat is great though so..
yeah, I have one where it did some of the new lyrics, with a new vocalist, but went back to the old lyrics/vocalist
I think we need finer control over what we want out of it in regards to melody, vocalist and lyrics
for my use case, I'd keep melody basically the same (Covers does do a good job at making a new intro if the original's is lacking), vocalist the same, but high strength on the new lyrics, as they still fit the current melody
wow, this can be used as a "clean up" method as well @cunning mulch !
Original: https://suno.com/song/0b6021d2-534e-475e-945c-7b3293c1a97e
New Ballad version: https://suno.com/song/0da4b062-8406-407a-9978-7b668dbd5caf
japanese pop rock song. Listen and make your own with Suno.
japanese ballad, romance song. Listen and make your own with Suno.
Sounding MUCH clearer now
I'm trying to
I'm sure this was answered a zillion times but why seems that Suno tends to arpegiate a lot?
You can make both play at a same, time, and hear the difference as well. So once, you find the gernes that work well, together, it's really amazing how well it works
In what manners/places?
There are sublte variantions, and some creativity methods
I know it took me a little bit to get instrumentals tamed down a bit
The electronic song "Pulse of Light" combines a House rhythm with Techno elements at 155 BPM in E Minor, Dril trap arab song. Listen and make your own with Suno.
oh yeah... the contest. Does Cover also count?
Arab trap House and Techno at 125 BPM in E Minor, creating an energetic, mysterious vibe. Remixing Dril song. Listen and make your own with Suno.
Mostly in electronic music, but after a lot of generations I noticed it tends to make most of the time basic arpeggio melodies, like, you know, a basic arpeggiator
Which genres are you using? Suno has a tendency of acting strangely with certain genres.
ah. I generally work with metal, so in my case, it was obnoxious guitar solos
Arab trap House and Techno at 125 BPM in E Minor, creating an energetic, mysterious vibe. Remixing Dril song. Listen and make your own with Suno.
Well - i'm trying to do ancient greek. The AI is using modern pronunciation and losing the poetic meter
oh...
House and subgenres, Synthwave, Future Funk...that kind of music, even when making piano solo music or whatever I feel like to prompting it tends to create basic arpeggiator patterns
Are you able to do phonetic replacements to get the ancient sound?
Yes, please do keep in mind, that the languages which Suno uses, are mostly the "standard" dialects, for example, if I use Swedish, it's a accent from Stockholm, and if I use emo or alt-rock it's an accent from one of the major tri-states areas. So there's really nothing you can do too change this, sadly.
https://suno.com/song/963b3652-4ff4-4902-a471-3a1cdb8519fb here's a test. Phonetic replacements would definitely work. i will try
Traditional Greek, Folk, myth song. Listen and make your own with Suno.
What does this mean?
I'd need an example of a word that's being mis-spoken
i'm asking the ancient greek discord and they're saying basically all of it.
"Since vowel length distinctions are wholly dispensed with, losing the very basis for poetic meter of the time"
Yea, so it goes back to what I informed you.
#📚┃suno-school message
yeah
I'm not sure, if there's anything "we" can do too change this, sadly.
At least with English, you can type a word phonetically if it has problems with it, or substitute it for another for the vocal result you want
Yeah, I think that would be the best approach
like, I had a song with 'caste' in it, and suno said it 'cast-tay', so I dropped the e in the lyrics box
well - it's just for research. like..i'm pretty sure no one is singing the illiad with the right pronunciation and with the poetry meter being distinctly heard. it's a big test to see if AI can do it
oh i think you might have trouble with that.
and these days, I'll never use 'our' in a song... too ambiguous.
and it'd be pretty cool if we could hear a 2000+ year old "rhyme or beat" from the dead language/poem
there's also Latin
I wanted to make a song with a Zen sutra as the lyrics, Suno did not understand most of the kanji, so I needed to rewrite the whole thing in Katakana which is basically phonetic japanese. if you are creative with mispelling the words in the right way, I think you might get lucky with ancient Greek
Original: https://suno.com/song/13fa7210-950e-4377-a088-fb105073dfc1
New: https://suno.com/song/47ef3bd1-897f-4653-89f7-ebc92b6b7630
reverberated, uplifting, tropical, female voice, calm song. Listen and make your own with Suno.
Wonder why it didn't understand them
@graceful condor I'm listening to the first one
Could you try, to add more Style of Music?
And narrow it down more? How many characters are using when you are prompting?
theyre archaic kanji... most modern people dont know how to read them properly either
Ah.. left out of the reform
Those were just examples I made now but as I say it tends to do that a lot, using more or less words on the prompt
I have an idea...
Sometimes I tried to force it to syncopate or whatever using the word syncopate trying to get a nice melody / song
Yes, but what I've found, at least with Suno, is that it doesn't like to use few prompts, and the range can get quite large of what it likes to use. For me, I have to prompt more arpeggio on my songs. But I stick with emo, pop, and alt-rock, it could be that House has more Arpeggiation associated with it.
what if all languages used for suno adopted the IPA as a backbone?https://en.wikipedia.org/wiki/International_Phonetic_Alphabet
Here's an interesting Cover. It 'interleaves' my fixed lyrics... https://suno.com/song/143242b1-80e2-4851-b3fe-8a39143c6573
Well its weird because not so long ago I made an ''album'' of instrumental stoner rock and it happened the same it tended to arpeggiate a lot even on stoner rock tracks, with guitars going scale up and scale down, I had to discard like 100 generations to get 12 or so ''good'' tracks
What are you prompting for the instrumentals?
I've found that if you call it the [Hook], and use tags like [Extended Hook] and [instrumental Hook], you tone down the 'solo effect'
See, I would LOVE to have an exception tag for Suno, to prevent this from happening.
just negative prompting in general
ah dude
i got another AI to transpose the ancient greek to IPA and tried to have suno use that as is...
i'll let you guys know if this works hahaaa
https://suno.com/song/b98fa6f3-edf3-45a3-b043-6de0472f225d I think it sounds great now... lol
Traditional Greek, Folk, myth song. Listen and make your own with Suno.
#1253341197917884530 message
Might as well upvote this one: @cunning mulch , @graceful condor
I'm using the Cover feature, using the same lyrics but changing the style tags. The results either sound the exact same as the original song or they're a totally different song, just with the same lyrics.
I'm doing something wrong?
for now, it seems best for attempting to replace lyrics in-place
That's a nice feature, even if it's not the intended function
I was envisioning something like when you see slow acoustic versions of 80s songs
Perhaps trying to convert metal into a soppy piano ballad is asking way too much.. 
Yea, me too...
I got a 3:37 just now
symphonic metal song. Listen and make your own with Suno.
and some lyrics that are not in either of the songs

It's always a risk too watch these songs, I'm in a ballad mood, and I watch this, hahahha My brain woke up
Sorry. 
haha, you're good!
Now, it would only be fair of me, too share my song, naturally 
Do it in showcase and mention me. 🙂
Ooh, sure!
What happened?
No, it vanished
love ballad, duet, male and female singer, duo, romantic song, serene song. Listen and make your own with Suno.
There we go
I just didn't want to take this off topic but w/e
What was this before?
Wait I should just check
interesting. You can influence a cover some by editing Displayed Lyrics for the song you're covering
I went back and tried it on an older candidate for Springfield Cleaning/Air Mail Savages, and while I was able to get the reprise that the final on HonkFM has, it did not pick up the wording change in the middle
Feel like it's just gone, "Taylor Swift, got it"
Oh? I don't listen too her, is that a song of hers?
No, it just sounds like her
That's a great compliment, thank you ❤️
I try to use up over 3000 characters on all my songs
I might be referencing completely the wrong person because I get all those kinds of people muddled up, haha
Do you write your own lyrics or use the bot's lyrics?
But that doesn't impact how the song flows, right?
I write my own ❤️
It depends on the scope of your edits
wow, fascinating, didn't know that!
Cool, me too. Though, that's fairly obvious in my case. Haha
"It's what separates us"
I used the bot scripts at the start, but it all felt like: " This is an AI song " . And I hear so many of them online now, that it's obvious
Yeah.
They also just don't make sense half the time.
yeah. I felt a huge difference between even cleaned up AI lyrics and finally writing my own from scratch
Exactly
can't share the human-written one here (too spicy)
12 tokenzations now, still the same sounding, so strange this cover system sometimes.
tokenizations? Explain?
Like, "generation" "new song"
oh
I guess it doesn't want to leave the korean version, over to ballads, I reckon..
it's giving weights in different places
wow, it's just not working at all with: korean, rap, k-pop, bubblegum pop
Try using strange words or invented ones, sometimes the AI gets creative that way
That's how a lot of mine end up. Haha
Imagine if we could manipulate that
I know, right?
hip-hop, soul, r&b works though
Hey, my song is on infinite loop, cannot download it, doesn't appear in my library. Any solution?
I'm getting some 1980's anime style now, fascinating.
Gonna pop back into Impractical Ideals v1 and see if I can get those lyrics updates to show
It's on infinite loop?
Have you tried playing a different song then going back to it? The song needs time to calculate its length etc
Oh lol, now after playing a different song it transformed into something quite different o.O glad I recorded it with loopback mix.
That's really weird...
It went from being delicate horror ambient to being house music xD
Magic!
hmmm.... I can tell when it's incorporating lyrics changes
I'm having loads of success with just one of my tracks.. like it's producing some great stuff.. But other tracks it just hates it
A song. Listen and make your own with Suno.
also, the art generator even sees it as a hippie rug
Lol the song is back, the suno generated x3 times the same song
So, some feedback on Covers: Suno will very rarely choose to honor any changes in the Lyrics field without also altering cadence and melody. If it honors melody, it's 99.9999% going to choose the original lyrics
Guys, any tips on extending songs to make around 4 parts for 4 minutes each? Way too often extended ends up being 10s up to 70s.
I would like to see perhaps a radio checkbox to choose which we're most wanting to change, rhythm, melody, lyrics, etc., and have Suno give more weight to that choice
I do, however, like that it will generate a new intro, which can help a song with one that's a bit lacking
Where can I find liked songs I liked by other creators?
Covers: Just some feedback for Devs: It REALLY REALLY wants to slow everything to Dwarven Chant levels. I wish I could upload a whole song that I made on another site (though I can understand not having that ability), I'll get the hang of it. Though i've wasted most of the first 20 covers because it just keeps slowing everything down terribly (Same tags as original song too, not switching genres), and then it'll switch genres half way through the track.
Wish to see in general: Tiberious has a good idea there, too but I would absolutely ADORE being able to reuse seeds. I work through songs in a strange way probbaly but i like the sound of one but mess up the lyrics, then using the same seed would be greatly helpful.
https://i.imgur.com/O2f0voJ.png
So this keeps happening. I found a track I wanted to cover (Generated fully on Suno), I try to change the genre tags (Everything else is the same plus the cover) and it does this. It's also not anything above G rating
try asking chatGPT if there is anything inappropriate with the lyrics, apparently suno uses that for content moderation
it's literally a century old hymn
ok, you can still ask
welp
lol. just looks like a bug
because I generated those lyrics earlier and thats the track i'm trying to 'cover'
oops.. I thought this was bugs channel.. headdesk
Errr.... that's "bohemian rug" to us more refined hippies. Lol
Dang autocorrect!
Hey. I just thought. Inferential leap. If I speed the pitch and save so a 2 minute song fits into 59 seconds... upload it to the audio, and prompt for it to extend at 20BPM., will I get a new song at the right tempo. , he said laughingly.
Hi everyone,
Noob needs some help ... 😬
I produce music and I want suno to get me some basic ideas and musical elements like melodies, grooves and basslines for intros, build ups and drops.
I've blown about 3.5 k credits already with veeery few useful results 😦
Let's say I want to create the intro for a dark house banger with 128 bpm in the key of D minor with:
- Dark, distorted ambient texture that turns cleaner towards the buildup
- slightly distorted pluck playing a simple eerie yet uplifting melody
- highpassed drums and Percussion
- simple reese bassline
How do you guys get specific to get the sound/vibe you want without using more than 120 characters? How would you write a prompt like that?
I tried adding one part after another using extensions but didn't work at all . I also tried putting an artists name in but suno says gives me a 🖕when it detects an artist name as you know...
Also I feel like I'm always getting the same sounds...
How do you guys tell suno if you want a specific bass sound for example?
Thank you guys in advance
Well, I get an intriguing result every now and then, but for me, it is purely random. I rarely see any connection between my prompt and what came out, lol. Try experimenting and even including non-music-related phrases. My "best" content has come from NONSENSICAL prompts - which were also the LEAST GLASSY compared to the USUAL GARBAGE Suno makes (which I can't listen to - because it leaves a horrid "glassy echo" in my head for an hour or so).
Best results come from composing a piece in your DAW sounding exactly as you want it and uploading then extending from there. It can still be hit and miss!
Most prompt instructions are quite hit and miss and not very well understood. Key is hit and miss, BPM is hit and miss.
My take on AI music is that I view it all as experimental and I love it for its idiosyncrasies. I enjoy composing short pieces, uploading, mixing genres and hope for something really good, which I mostly get.
Its not at the stage yet where you can compare it to composing, mixing and mastering in a studio, but the time will come, I'm sure.
Will Cover feature be for free member?
Not yet
But like will it happened at one point?
Perhaps after the Beta stage it might happen
I see. I hope I can use it one day.
It's really cool, but still in testing stage. Hopefully devs will release it for all once testing is done.
covers don't work for me. it always says "a clip got stuck while generating" or something
The new "cover" feature is great but I find the upload time (max 60 sec) not meeting the excellent option to get AI fueled remixes of my own full tracks. Can the restrictions be lifted? And another thing, some (general) generations lack proper intros, can the "extend" option be optionally directed to the front of a track?
How's this, Woodhouse?
Suno Covers - Tips, Tricks, and Trial & Error
I'm gonna be exploring this new Cover beta feature quite extensively, and thought it would be useful to have all my findings in one place.
Suggested Experiments
The potential of this feature is hard to quantify right... but here's some things we can start with as we work to understand just how powerful it is:
The Covremaster™ :: Recreate low quality songs in better quality without losing the things you love about the originals. (PROVEN)
The Coveraoke™ :: Create Instrumental versions of existing songs (i.e. rather than removing vocals, re-generating the melody and music without them) (IMPLAUSIBLE)
The Acovpella™ :: Create acapella versions (i.e. just like the previous suggestion, but reverse it) (PLAUSIBLE)
...and many more to come I'm sure.
Tinda's Experiments, Phase One
First, I'm going to be attempting to remaster the very first song I ever generated (using v3), which made its way onto my first album. I love it, but I have never been happy with the quality, especially as it pertains to sound degradation over the course of a 3:45 song. If I had it to do over, I wouldn't have been happy with this and kept going, but it's been in my life for 5 months now and I have a certain affection for it. So, if I can somehow get a "re-do" and coax a higher quality version of the same song with this feature I'm gonna be real pleased.
The song: "Party of One"
Initial Generation: [1] (see links below)
Final Full Version: [2]
Initial thoughts:
+ Suno really wants to change the pitch, even if it's adhering rather closely to the original result.
+ Looks like even if you do a full 3:45 song, Suno will only look at the first 2 minutes. This isn't unexpected, but still disappointing. My first plan will be to retrace my steps, so to speak, by covering the original generation, and then covering any and all extensions I did on that, and then I'll need to re-assemble them in Audition afterward. I don't mind this extra work if it results in a higher quality version.
Initial result: [3] (from the FULL version)
- My goal is absolutely achievable. The quality still isn't perfect, but it's a substantial improvement. And getting a better one is just a matter of being persistent.
- The 2 minute limit is definitely an issue as this clip shows when it veers fully into its own thing at around 2:05, but I've got ideas for that.
Experiment 2: "Fenchurch"
+ The final song on my fourth album is just one voice and one piano. The initial generation was so good that I was content to overlook some quality issues because of its strong emotional resonance and Suno's beautiful interpretation of my words[4].
Here's my first attempt at a full Covremaster: [5]
- ✂ Process: I cut up the original version into 60sec overlapping segments and uploaded those to Suno.
What do I mean by overlapping?
1st clip: 0:00 - 1:00
2nd clip: 0:50 - 1:50
3rd clip: 1:40 - 2:40
And so on. This will allow important wiggle room when you are splicing the news segments together later on. it'll save you major headaches.
Next I created "Covers" for each of the clips (between 4 and 10 generations for each one to find the right sound) to ensure the separate clips would match up when I re-edited them together in Audition afterward. It took about 40 of my free Cover credits to accomplish everything.
Links
[1] https://suno.com/song/7fc942e6-47e0-4555-92ec-c4e4681d1276
[2] https://suno.com/song/04d379a4-fc09-475d-9c00-cea11e00bd23
[3] https://suno.com/song/82d887fa-8b1f-4e7f-b7bf-b69a495a6b65
[4] "Fenchurch" Original master https://app.landr.com/projects/1d763581-4a58-4b36-980d-177ebba183f6?publicToken=4d737178-676c-4820-9c07-c4122ef24836&sharing
[5] "Fenchurch" Covremaster: https://app.landr.com/projects/637cdeea-47d3-4b83-840a-759cef2442e8?publicToken=99636ff2-31e6-42d7-a41c-a22481ec96cb&sharing
Sure, just put 'em in there. Most of the time that's all you need.
Sometimes you can do a warmup like... I find whoa-oh! to be quite reliable, and then you could just do whoa-oh! hey! go! Suno! and you'll get some nice interpretations. For screams and laughter it's less predictable, but tags like [screaming] tend to work the more simple they are.
Now, if you're wanting to generate audience noise that's a separate issue and you'll waste a lot of time and credits trying to get what you want. Better to add those in after in your audio editor of choice.
But whoever presented Suno to you this way was mistaken. Though I suppose it depends on what kind of song you like.
Browse around the other instrumental stuff on the site. You'll see lots of creative ways to utilize the lyrics box that results in instrumental generations.
I see so we're using the corporate definition of "songs you like" then.
the cover funfction generates much better audio quality sound..and even compositions are better. I don't know what they did, but the audio quality is much better! I made some remixes you think they are CD quality
Well, it can for sure. But not always.
I don't understand. The vocals change the instrumentals? If so, yes.
I'm not sure that's true with Suno, but everybody's got their own approach. Best advice anyone can give you is just try stuff and see what happens.
when should we actually press the thumbs down button after a "bad song" was generated by the cover?
I wont make them public...
Looking at the announcement... I'd thumbs down anything that is a repeat of the original song, though I would also do this if it is too different (sometimes it does generate a completely different song)
It likely flags it for the developers to see. I would not expect it to need to be public.
but repeating the song is needed at least one time! Cause it redos the song and it has better sound quality...which is neeed. Here mabye a prompt option would be needed to achieve a remake of the original if you want.
As far as I'm aware it gives feedback to the devs in any case, for beta stuff like this only. It's unlikely to be automatic or require it to be public.
yeah, it makes sense
I want them to add a "strength" control
Because I mainly use it to make small improvements
I am really strucked by the sound quality !
it sounds like 4.0 already...
but also the compositions and remixes are miraculous with cover... it is different than create! Like another ai ! I am impressed !
I remixed some Udio songs ! It is AMAZING !
I cannot leave the PC anymore. Suno devs, I would kiss you for this feature !
I cant wait to remix my classicaly made songs, made with the daw, Studio one!
it makes also a huge difference to hear the songs in better audio quality
Please consider changing the maximal lenght for uploaded songs.. 1 min is sometimes too little. Why this limitation ? 1,5 min would be better
Limitation of v3 -- v3.5 has a 2m extension maximum 💪
There are big differences, in the two, so I understand when artists prefer to use one over the other;
- v3 is more willing to create a very wide range of genres and sounds
- v3.5 made some sacrifices to gain phenomenal strengths in regards to how it can "tighten up" a song, or help the lyrics adhere better to the song.
As long as you have started off strongly with a melody for it hook onto, I find that v3.5 really shines.
I'm excited to hear what kind of masterpieces you put together, my friend 🤘
my last song is too spicy to share here, but it had two previous versions I was okay with before being wowed by the 3rd (which basically came all in one shot). This was before Covers. Now, I have cleaned up versions of those first two takes
I do wish it was a little better at letting you reliably make minor alterations to lyrics in-place, especially without messing with rhythm or melody
I have a quick question, regarding the cover feature. Is there a minimum length the sample needs to be to be covered? Every time I try to cover custom voice clips it gets an error
The error was "something got stuck. No credits taken"
are you even able to cover an upload?
I think you'd have to generate something in order to Cover it, right?
It just let's me create a cover for an upload
Idk
So would I need to upload -> extend and then cover it or?
But that would defeat the point of covering no?
If it would just make a new song alltogether first
hello everyone. how to activate a promo code?
2 bummers to share;
- There aren't any promotional codes 😦
- Suno doesn't even have a spot to enter a promotional code.
wow.... Okay, got it. thanks
Could you give an example of such a nonsensical prompt and maybe a link to a song you created with that? 🙃
Can we pick a voice generation previously used and be able to stick with it yet?

