#πβsuno-school
1 messages Β· Page 103 of 1
Does the issue persist after you update the browser?
I already update, it didn't help
i was just getting an error when i double tapped
I tapped just one time
Tried extending it from 00:14?
How can I connect 2 songs thematically?
I want to make a Space- Opera.
I made Part 1. I really liked.
Now I want to make Part 2.
What are good options?
Had the same issue when trying it with Opera on my phone, I have uninstalled it and then via Chrome it worked
I cant get rid of the "high electronic plinging" It is so annoying!! https://suno.com/song/cd720fc2-4987-4273-afe5-5402f39936d9
Female vocals, emo, avant-garde, Vogue, Epic Chorus song. Listen and make your own with Suno.
90% of all my try's are like this.. Cant use any of it.
@glossy shard -- did you check the server for any information on this before making your post? To see if others are reporting anything similar, or if there'd been any official statements on the matter?
Oh, hold on i might have missed that.
Ah there.. Okay well guess ill wait then. wasted like 1500p trying to get rid of it
Thanks
I'm hoping there'll be a way to fix up existing songs when it's resolved.
So worth holding on to anything that otherwise sounds good
There's a couple different approaches the community is uncovering of minimizing the effect, but there' differing opinions on the results -- like furrybeans, I'm excited to see how things will play out
So again,
What are good options to connect songs thematically?
- When I already made 1 song
If you're on a paid membership, you can create a persona from the song so Suno will attempt to use the same instrumentation and vocal styles in other songs. Otherwise you could extend from the song and turn it into a longer composition?
Is there a good way to promt or an easy way in post production to get rid of that "clickling noise" that tends to happen on more than 50% of my songs since version 4 was released
What you mention is Shimmer and it has been discussed and mentioned serveral times (every day) here and in other threads. It is an issue specially present in most V4 renders. Devs are aware and working on a fix. So patience is your friend for now. If you still want to produce tracks, I recommend sticking with V3.5 for now and save the ones you like for remixing in V4 once the shimmer is fixed. π
And yes you can fix it in post (sometimes) but it requires good software + patience and skills π
well i mean i have done a few and removed the "shimmer" in my DAW it just takes soooooo fing long if you wanna get it removed good! So i was wondering if there was like a quick fix someone had figured out.
Like thinking the shimmer itself is like a "instrument" in the AI that creates that sound and you could ask for song with like promtion "No cymbal drums" as an example to get rid of it. Or if there was a super quick method someone has found to get rid of it in post.
I have managed to do it on all song i wanted to do it with so far, just adds so much work and i am SUPER LAZY
@icy kindle Is there any chance for opening up for making Upload audio extending work again for v3.5. until the v4shimmer issue is resolved? π
Heya. anyone know a fix for this issue? --- I really like the way a song is sang, but i want to change the lyrics a bit, but when I 'Create a Cover' or add a Persona made from that song, the words get all garbled and make no sense
Not that ive heard of, and its not from the lack of ppl trying.. But its a reverb(ish) issue so tracks with little to no reverb in its flow seems to have less shimmer.. i.e. rap has little
you can inpaint that section of the song to change it! π
ooo that's a thing now?
yes its not flawless exactly, but give it a few tries and generations and i eventually it works good!
go to edit then "replace section" on the song you wanna fix! π
Thanks skytrax
No worries! π
anyone get suno to work on android auto?
Getting a 404 when signing into Google? Change your default browser.
I had that issue with Opera and switching to Firefox fixed it
no, just audio doesn't play
Ah π«
Have this issue were I cant really fin a good workaround. I want the entire track vocal to be Female vocal and the same vocal. However if I get too creative with [] prompts in the lyric window it tends to always switch to male vocal for some verses mid song. Things like this [Bridge - Ethereal and powerful with lush strings, rising brass, and dynamic transitions]will often result in a vocal switch even if Female Vocal is first in the style window. I have tried to make sure that the in lyric commands do not have the word "vocal" or "voice" in them not to confuse the AI, but it often still makes gender swaps mid song. Only "fix" i know of is keeping the in lyric commands extremly simple, like [Intro] [Verse] [Chorus] etc, but it seems a bit limiting for building a unique song. So Anyone have a good fix/workaround? π
add "female" to the beginning of the prompt?
[Bridge - Ethereal and powerful with lush strings, rising brass, and dynamic transitions, Female Vocals]
tried and does nothing π¦
[long description of thing you want]
[female vocals]
?
holy crap with the damn shimmering in V4 . almost unusable 4 from 5 songs are baaaaaaad .. wtf
Believe I also tried that some time ago with no effect, but will give it a go again. Also 3 commands in a row wont work.
How much of the longer description is followed?
Just wondering if something in that is pushing it the wrong way
lol try making 1500 K credits worth of tracks just to throw them all in the bin due to shimmer π Stay with 3.5 until V4 actually works, then remaster your tracks then π
ive at least 3000 k shimmering
This is probably not related, but I'd try being less verbose in those tags. I suspect a lot of those words are just ending up as prompt noise with Suno cherry-picking certain ones.
I mean I got different vocalists by specifying household objects as the sole thing in a metatag before so there's a chance it sees "powerful" and go "aha! That means we need a different voice"
it takes some time, but you can generate without it. make it in 3.5 and remaster until it goes away
and credits π
#πβsuno-school message it's not nearly as bad as you might think though
My credit vallet thinks otherwise, but happy for you that youre not rendering loads of shimmer tracks π
those were failed, only 4 of them, so 40 credits till a good one isn't horrible in my book
My workflow.
Bridge: (Ethereal and powerful with lush strings, rising brass, and dynamic transitions, Female Vocals) works 80% of time
@crude jetty
Suspect it can be individual on the frequncy depending on the styles one tend to render, some genres seem to be more prone to it. my shimmer percentage is roughly 85-95 % I would say. though im not using V4 for now so dont have any update for today
Have you found "dynamic transitions", "lush strings" and "rising brass" to be working consistently in the past gens or in this one, before tacking the gender vocals?
Sorry, that's an open question. I do folk, pop and style similar and always get a consistent female voice all the way through. Possibly by the minimalistic styles presented in those genres.
I make grunge songs, trust me shimmer was horrible before
Nope, not always, but for me atleast it tends to throw some interresting things into the individual tracks and sometimes it hits a home run out of say 10 renders. Thats why i like to use it π
ok, not making anything in that genre so can't say. But for reggae, Pop, Soul, etc its bad.. like VERY bad
it used to be miserable in grunge too
Thanks a lot. Doing it by hand wouldnβt be a problem, but I created one version without personalization as a template, and then tried to create others including different names. However, it didnβt insert the names at all. After ten attempts, I stopped and wanted to hear about your experiences. Thanks for your reply! ππ»
Great. Have you tried to use a persona instead. Or make a persona from what you already have, then run the prompts again and attach the same vocal persona, if you like that one, or use another.
also , if you make your own music , like i do .. i like to upload my songs and let the ai create new variations .. the outcomes are very cool , mostly surprising how good it is .
Yes, but seems to bug out there as well, but will test various things now and report back if I find a good work around, just need to set up a good testing ground / system π
Yup, thats what i Use it for mostly, work on some things in midi and with vocal and let Suno make magic with it. π
i had sooo high hopes for v4 .. but .......
Still Early days, It will get better.. times is our friend here i guess π
3,5 sucks for mushy voices , and v4 sucks at all lol
As for work around, as lyrics are often 1/5 of your vibe, run a minimal gen with just lyrics and any prompt, kazoo, piano, triangle, whatever. Keep lyrics and metatags as you wish, maybe it will pick up on that as well. But id do one with lyrics as is, one with lyrics but without the metatags....Get it to just do female all the way through but pick up a bit of style from the lyrics vibe.
Make a persona from that result.
Reuse prompts of your original so lyrics and metatags are all in their place again. Select and highlight ,, and copy your style prompts.
Link your new persona (which will replace your prompts)
If style prompts are now different from original, then select and paste in your original style prompts.
Run and see what you get.
This song pulses with vibrant energy and irresistible allure, capturing a magnetic love shrouded in mystery. Feel the provocative rhythms and Caribbean-inspired magic as it questions: Is it love, or an enchanting spell? Click to feel the beat.
What's considered a mushy voice?
Hello Suno people. I am in the early stages of creating a "Suno Song" redeem on my twitch channel. I have a commercial license, and I want chat to be able to put in a prompt and then once the song is ready, I intend to have it auto play on stream, with subtitles and all. I am aware its on me to balance the chat cost vs my available credits on Suno. So my question is, has anyone gone down this path before? is there an API or would web scrapping be my go to on this one. Basically am I starting from scratch, or are there foundational pieces I could work off of? Any info anyone has would be appreciated
I am a coder, just wanted to make sure I wasn't working on something that already existed in some form or part
Also generally if you're trying to automate song creation via any form of API that's very likely going to be against ToS
Twitch ToS or Suno ToS?
Suno
well then, that's unfortunate, seeing as I pay for it, and have credits to use as I please, seems silly that automating a process that I could done by hand the same way and be allowed. So you are saying if I did this by having chat post their prompt, and I pause my game, alt tab, paste it into Suno and then use the song... that would be fine for ToS... but if my bot pastes it, thats a ToS violation?
^but if my bot clicks the download button instead of me, thats an issue?
not trying to be snarky, trying to understand the difference here
As far as I'm aware any automation isn't really allowed. Whether it gets detected or acted upon is another matter. For example "rabbit r1" interacts with the web page and isn't a problem, though not officially "approved" as such.
The definite no-no is using anything to directly access URLs to automate generation. That will lead to account restrictions if not bans.
fair enough, if thats the ToS, then I shall abide and find another platform that will allow my commercial license to be used as I need it to be. Thanks everyone for the quick and thorough responses. Take Care
Thanks π
When Suno makes improvements, will they generally make a new model, like a 4.5, or will we just, for example, start seeing less shimmer in 4? I'm just curious because my V4 creations have been sounding a little better in the last couple of days than in the previous week, and I wonder if I'm lucky, or Suno is pulling strings lol.
For anything in beta (v4, covers, personas) things can change. So yes, v4 could suddenly improve.
Once it's out of beta it may be possible to make small adjustments outside of the model itself, but anything requiring model changes would have to be a new version (e.g. 4.1 or 4.5 or 5 or so)
Ye, in beta things usually don't change version numbers, at least not major version number.
It will most likely stay at v4 untill it's released, but who knows xD
Yea I think you would have to disclose that once one submit a prompt, that output was owned by "me" the twitch channel operator. You won't get many takers then.
I have a hunch they've hired some real musicians and vocalists and , while they work out their challenges, the band has been making our songs for us. Keeps things flowing. π
Yes, that's how you do it, when someone creates a song they accept that they do not own it and have no right to claim ownership of the generated song and the lyrics.
Should be fine then
I think I stepped outside of what was asked here. The OP just wants to stream outputed songs from people randomly submitting ideas. I suppose there is no file sharing or downloading uminvolved.... so if someone wants the song, they'd have to record it while its streaming.
So, who knows. I mean, if you just sat there, as some have already done, and just type in requests for folks and play the output. How's it working out for those folks?
Ye, so they redeem it and it auto generates a song and plays it in stream, no download etc for the user, that's what I understood from the OP.
In that case it would be the owner of the accounts song no matter who created the lyrics i'm pretty sure.
If it were to be copyrighted lyrics from a real song that might be something else tho, could however be seen as a cover and fair use in that case as well.
I assume that would be the person how owns the accunts song at that point as well.
Don't quote me on this as I'm not sure, but I assume that this is how it would be.
wow, appreciate all the talk about this, I decided that the only fair thing to do before cancelling and moving on, was to email the Suno legal team for ToS clarification. So I have done this and am waiting for an official response. As for the "ownership" issue everyone is back and forth about, Twitch ToS says once you say it in chat, you don't own it. There is no expectation of ownership when you say something publicly in twitch chat. This would mean that ownership is a non issue
However ToS of suno says that if someone else wrote your lyrics they own the lyrics and you the complete song but you need their permission to use the lyrics.
So this is interesting to say the least.
Suno is best suited for making new music with new lyrics. If you're working with another writer, you must obtain permission for any lyrics or other ...
I just found this btw, so I know I'm talking against my earlier message π
But they do mention specifically writer, so I don't know if it actually is a thing here
that says lyrics.... the users wouldn't be providing lyrics, they'd be providing prompts and Suno would generate the lyrics. I notice that link doens't say anything about lyrics prompts
But the prompt you send in is the lyrics no?
Or will the user only provde the "Style of music"
well like the user could say "jazz song about beating a mario game" ... I don't think that qualifies as "writting lyrics"
exactly that yes
Hi guys, does v4 function correctly?
Simple answer, no
Less simple answer... yes and no π
But i would say more close to no right now
something seems to be going on..I just tried a couple of my most shimmer-prone prompts, and I got like 6 or 7 out of 8 songs back that were essentially shimmer free at a quick listen. some of them sounded filtered, with subdued shimmer, some i didnt hear any telltale shimmer sounds at all.. i hope it's not just coincidence.
That's good tho, maybe they are doing something, however I really hope the songs wont get filtered down to, I like the sounds now, it's perfect as it is just no shimmer π
Good luck, I like that idea, I'm a coder myself, I do stuff like this all the time for all kinds of thing, I hope it works out, would be cool!
can you define for me what the persona feature is?
Basically copies the voice and instruments from the other song so you kind of have the same person singing.
So if you want to make an online singer person t.ex you can have the same person sing in all your generated songs.
That's how I have understood it at least.
so it is primarily a voice style tool?
it isn't a style derived from an uploaded song?
I don't think it's meant to copy the style but mostly voices.
However I do know it copies the "Style Of Music" when used so I could be wrong here.
I don't have to much knoweldge of it, so I recommend to research it a bit deeper and see or someone with more knowledge then me can answer possibly
I am wondering if it has to do with feeding it songs, and then having it derive a template to create derivative works of a similar style and sound
more than 20 shimmer free songs now. hmmmmm
What's your secret? And are these remasters or new?
both. just did a remaster and it had greatly reduced/no shimmer compared to the original. and most of all my songs come back shimmer free. I don't know what's going on. Secretly 3.x under the hood for me?!
Did a couple 1 hr ago still shimmer, might try again l8 π
Lmao damn shimmer destroying everything. It's like a disease
feel your pain, one thing is the (Just render 15 songs and you will get 3 shimmerfree ones) advice, doesnt help if the 2 banger ones are shimmerfests.. But im sure its on the top of the list for the team though. π
You choose customer, it will can write lyrics
That's why I asked my question from earlier about V4 improvements. I've been having fun these last couple days. I'm tempted to remaster some really shimmery songs and see what happens.
Would you see if this will help you fix Personas please. It's not spam. It's math.
Shimmer seems better for me this morning.
Cymbals sound like cymbals.
put it where?
shimmer is better but the remasters are actually just covers of the original song, they sound way to different
One fix at a time. The shimmer was the worst issue.
We have a "workspace " area now. Woo woo
There's no backend focused one, so I usually pick suggestions-browser
very silly question that has probably been answered before but I cannot find this info anywhere. What is the context length for Suno? Is it model specific or is it the same regardless of version?
They definitely worked on the v4 shimmer. Some great, clean generations this morning.
stop it you might tempt me at giving it a go? π
I've done at least 30 generations today in v4 with zero annoying shimmer. Granted, I've only done bluegrass/folk but these were getting shimmer before.
I don't think this is officially documented anywhere... I think it's a couple of minutes for v3.5/v4 but I may be wrong.
I'm going to do some rock/grunge remasters to really test it.
I feel like it's less than 2 minutes unfortunately
I want to use the mobile app for android but i dont like that they share so much info with 3rd parties. There are plenty of apps that require no permissions. why the invasive permissions? feel like its always the same reason, the almighty dollar, Make a buck to exploit people so they can sell user data to 3rd parties? its not what i signed up for. Very disappointed as i was looking forward to using the mobile app.
I have a problem... please help. With the release of the mobile version of SUNO on Android, I was faced with a very nasty situation. All the songs I create now only indicate that the application is fake.
And so is every song created, whether from a phone or from a computer.
I'll try to contact them.
If you're sure that you downloaded and are trying the official Suno app and not one of the imposters, fill out and submit this official form from Suno. It's regarding that very issue. https://sunomusic.typeform.com/to/zGsJSzsf
Turn data collection into an experience with Typeform. Create beautiful online forms, surveys, quizzes, and so much more. Try it for FREE.
Is there a trick to getting personas to sing new lyrics. I wrote lyrics and liked what I heard so I created a persona and now when I change the lyrics and click create it continues to sing the old lyrics.
hey my suno is not using my audio anymore for the style of voices. it is just using generic AI voices now. I am using the extend feature and have done numerous songs with it. Did they change something?
That seems broken and behaving more like the Cover feature instead
Unless there is an announcement the majority of us know what you know when it comes to changes
Does anyone know if using * (Sound effect) * still works at all? I haven't been able to get any sound effects to work.
Maybe its just me, but just confirming what have been mentioned here earlier. Shimmer in V4 seems WAAAyyy better now. π
Still hear it at times in rock songs but not as often. Haven't tried the stems to see if they are there in stems yet.
Not hearing it much at all in more acoustic songs.
Really? It was in something I was reading that someone linked here about it.
Let me see if I can find it.
Yes there has been a way to TRY to get it to make something like that but it isn't supported
Hmm, it doesn't seem like it was official, darn.
That's definitely in the same page I was reading.
If you really want it you can try to get a sound out of it that's usable and then edit it in to a music track that didn't originally have the sound
But as for just asking for a sound effect inside of a song and having it just generate it? Not really possible
Not in a consistent way
Yeah, I was trying to see if I could get it to make a gunshot effect. But I could just add one in audacity or something.
Sometimes, according to that
So it needs to be in brackets instead of the asterisks then. I'll try it out.
I use allot of meta tags and find it is difficult to get any cowbell sound. i put in the description and metatags in the lyrics. i like drift phonk music and i tell it i want 808, 909, 404, 303 cowbell, aggressive drift phonk etc.. Its very rare that i actually get any cowbell rhythms, synth lines or riffs. it just doesn't want to do cowbell in versions 3.0 3.5 and 4.0
Anyone having issues with Persona's? They were working fine yesterday, but anytime I try to generate with one right now, they don''t stick to the generation
I'm noticing changes with the app layout... could be some things going up and down at the moment.
Uploads are handy for things like that. https://suno.com/song/f293b894-55d9-41a6-8066-926d256ad7e1
70's Rock, Blues, Folk, Cowbell, more Cow-Bell, Cow Bell, Guitar-Driven cowbell song. Listen and make your own with Suno.
LOL wellll shit. I made a great song and left (words like this) in the lyrics and well, she spoke them and its very stupid because its musical direction in the (). I can edit this out yea?
Put musical directions in []
You can try using the replace feature or start a new extension just before that point
yeah normally I do. I just didn't catch myself there.
Ahhh, okay.
its sooo funny man
song is great and then she whispers (Soft pads, fading echoes, a heartbeat-like rhythm)
and I laugh so hard I almost choke.
Suno isn't really that fussy regarding swapping [] and () unless the enclosed text is natural language, in which case Suno will be a bit more inclined to sing what's in the paretheses but will also often do the same for text in brackets.
ugh using replace is not letting me just grab that section.
I probably should have put emphasis on "try". Lol. It's... quirky ::cough::
LOL yeah I think I may have it
Hi all, I could use a bit of community advice for an annoying thing that keeps happening. When I extend a song, even when there are no lyrics in the box, it insists on repeating the chorus, or a part of the chorus, even if the chorus is already in the original song. I would move back the extension time if I knew beyond a shadow of a doubt this would happen, replacing the chorus that's already there, but I never quite know if it's going to do it or not. But, it's done it three times to me. Can you redo an extend but from a different time? Here's a full song I put together that has the problem so you can kinda see what needs fixing:
smooth, beats, rhythmic, r&b song. Listen and make your own with Suno.
Yeah, I know I haven't named the thing yet so that's weird too lol.
Have you tried extending as an instrumental?
... Why didn't I think of that? That might have helped end that particular song better. I have others that I needed to re-add a lyrical bridge that got cut off and it did the same thing with the chorus, but for that specific song that would have been amazing.
what time were you trying to extend the outro from?
I did 2:13, which was the end of the song.
You could also try when extending adding something like
" [Outro]
[Musical interlude]
$&^%&$&$^^$&#&$^$
[End] "
Or something like that when extending normally and not as instrumental.
I did try with [outro, instrumental] then [end] on a new line. What are those % and $ signs?
It's just to give it non-verbal input. Sometimes it works sometimes you get some weird interactions.
Huh, okay.
It looks like you're extending from midsentence since the vocals just get cut off there, so Suno's going to be inclined to continue singing. You'll probably need to go back further to just BEFORE that chorus, or do an extension to finish that chorus and then shoot for the outro.
It's "prompt noise" - basically anything that suno will process but not actually sing. I regularly use slashes for that
Why the f**k can't I sign in to your damn Android app? π‘π‘π‘π‘π‘
there is any member from support here?
i have the fake app bug for days and they doesnt answer on email
Yeah, that doesn't explain a thing.
but it's still a better place to post that question... π
especially since a dev posted there 10s ago, so your post may actually get seen by them
as opposed to this channel, where we talk about the technical aspects of creating tracks with suno.
So is suno looking into spiking neural networks
Cause that's gonna be the way, it's like we had no Internet, now we're at 56k dial up, and that's gonna be the dsl lol
Okay, that helps me understand why it continues to do that over and over. Thank you, I might be able to make some cleaner extensions with both of your help.
When i do a cover of a song with new lyrics the old ones mashes up with the new lyric. is it any fix for this? π
This is on your whole account?
when I see "an error occurred" its usually cause i'm trying to change the art icon and its too big. not saying that's what's happening here but I got nothing else to go on with this post.
Hello: I can't use "persona" and "upload audio" together....is it normal?
not sure if this is the right place, but if anyone can help I would appreciate it. I have invited a friend to suno and they have made 20 tracks but both of us haven't received a bonus, is there anyone we can speak to to fix this?
Anybody want to talk about Shimmer or what I'm calling flutter
Not sure what else there is to say about it that hasn't been thoroughly discussed in #1308499132356362383
π€·ββοΈ
Is there any tools available Psd plugins anything that can help with it? This song I just finished only has it whenever the bass guitar comes in and if you lower the bass frequencies it gets better but that's not a fix. It just seems like there's too much of it
some have had success with adobe audition, adaptive noise reduction with unsuno preset
There's a couple user-submitted threads in #1308499132356362383, but there's differing opinions on their efficacy, but I'd still recommend checking out what people are saying worked for them
Unfortunately there's no "one-size-fits-all" solution atm, since there effect seems to manifest with different intensities, where sometimes it might be an exaggeration of an existing sound, while other times it might straight up be an encoding artefact. Not to mention the fact that now we're more attuned to that particular noise, and are more likely to hear it when specifically looking for it (think of that douchebag friend who makes you REALLY listen to the live track because you can hear the amp noise on Hendrix's marshall stack -- sure, he might hear it, but people who aren't familiar with amp tones just hear fuzzy-noise)
So the Shimmer part of it is fine as long as it's not killing the song but then this one that I just made about halfway through it just decides hey I need to be louder and that's where the problem lies no matter how you do it. Even if you extend it prior to that point it will still happen
I've found the UnSuno doesn't work as well on v4.
Good evening, I have some kind of rattling sound in the background in almost every v4 song I made. Does anyone else have the same issue?
Shimmer. It is a known issue, being worked on
Okay thanks for the quick reply π
developers also dont know how there system work. no proper tutorial and workflow
Remastering a v4 song that has the line βShe sits at the barβ became βSheβs tits at the barβ.
For the life of me I cannot get a cover to do anything different than Remaster. It will not change genres.
I create a song that I like and then I say Cover and change the style to something really common like βhip hopβ and it plays a nearly identical song as the original. (In this case it was an indie folk style song, think Bon Iver).
It could be that the lyrics themselves are locking Suno into certain styles, which might be hard to break out of w/o tweaking the lyrics. Like your "I Did A Thing" track - I bet that one'd put up a fight against style changes.
I always see it as a way to inspire innovation. If we had instructions on how to use it then not as many people would experiment in ways the devs don't know or be able to think of.
basically all models that do.. partial latent rediffusion (for lack of a better term, audio2audio, Img2Img) have a parameter exposed (you'll find it on everything from Udio to Stable Diffusion) that goes from 0.05-1.0, representing the % of data destroyed in the original file, which is then filled back in by the model. for whatever reason suno has chosen to keep this parameter hidden from us, and they seem to randomize it. the songs where you're typing in hip hop on a folk song and getting folk songs are setting the parameter to like 0.1 and only destroying 10% of the data, and the ones where the genre changes significantly are lucky generations where the parameter is set to like 0.5-0.8. they're the only company I'm aware of that allows partial rediffusion as a paid service without actually allowing you to specify how similar you want something to be
big part of why I use suno for initial generation and then head to udio if I want control over modifications. tagging is a bit different but once you get used to it, using the two platforms together becomes a superpower
Wow. Well said. You know most people will not know at least 50% of what you said. You lost me and I have a little knowledge in the area. I knew about Bark using latent space generation, but not in the way you said everything. I agree on the weight part. We should be able to weight our prompts. Even if they don't tell us how, we should be able to use the normal weighting things at least like ((weight)) if nothing else.
I'm dealing with a preponderence of treble and hissing using v4. Anyone have a workaround?
(I assume that is what you are talking about)
It's not the shimmer though. I'll check out the guide, but it's like things like way too much cymbals in the percussion. Anyway, it's stuff I can't just correct in a DAW.
That sounds like shimmer to me, it is like a wavy reverb and very clearly heard/affects cymbals
Oh unless you mean more of a "loud" sound which results in... kind of like a busted stereo? Ugh, hard to describe that sound
Based on what you sent me, that's not my issue. Mine sounds tinny, like an old s***y AM radio. https://suno.com/song/1c2591dd-9ca7-488a-9bc7-b675b9875b8a
Minimalist, Cello, Cello, Cello, Powerful Emotive Clear Deep Male Vocals, Slow Build, Baroque Pop, Hip Hop Percussion, Acoustic, Melancholic, Melodic, Catchy song. Listen and make your own with Suno.
yeah this is bugging me too, lots of wasted credits
I'm wondering the same
Is there something wrong with Suno Video Regeneration? Mine never works. It times out every single time.
Interesting. Thanks for the detail. I guess it is a beta. I think that parameter could be exposed, but it would need to be framed in some sort of musical context. For example, can the model identify what the melody is? If I want to change an indie folk song to a hip hop song, can I control the parameter for melody, chords, and musical style differently? Maybe they havenβt released the ability to control this because theyβre hoping for controls that are more βmusicalβ!
If youβre hoping to take these songs and - without playing every part yourself - make them into something that at least sounds somewhat commercially viable, or you want to rearrange them, this is the way. At least for me. RX can clean up the artifacts / noise, and separate the stems. Pretty nice. Then into Logic it goes for mixing and mastering.
I'll try the base version first and if I get any success following that, then I'll consider shelling out for the top shelf packages
That's like, if I actually see a few dollarydoos back lol
No one that can do anything other than give suggestions. No devs or Suno employees in here.
Suno is more than likely slow today for multiple reasons.
There has been talk of fixes that were supposed to come in by today. I have no idea the validity of those words, but many have talked about it.
Regardless people keep assuming it is a fix for the shimmer. It is not.
You take all the people checking to see if the shimmer is fixed along with the Android app release and the offer of using v4 for free.
Equals slow Suno.
Just my thoughts.
I, for one, was expecting at least something - But not a word; I for one, will not be spending another dime on SUNO for the time being, until they fix the Shimmer. I've never seen a company who knows there is something wrong with their product, but act like it's the best thing ever, when by the amount of complaints prove that it isn't.
Ok then...lol
I know. Did you see me say I don't like doing that
I just want to know if anyone has updates
You spoke in different lines. We are just users like you.
Has anyone tried using some software to manually target and remve the shimmer?
If so, what did you have to look for?
Has anyone noticed hissing/high frequency sound on vocals lately? All songs I generated today have this problem but not last week.
Tried more than 100 times today and almost all of them started to hiss after 2 minutes
What do you mean I "spoke in different lines"?
When generating music, Suno started inserting some nonsense into the beginning and middle of each song.
https://suno.com/song/ded64210-c74e-43a0-90d2-6fdfcd14dc75
A song. Listen and make your own with Suno.
Sounds like your account was flagged for API usage.
Typical cause for the accounts being flagged:
- API usage (either by you or through Third-party services, including non-official apps)
- Specific IP address (e.g. VPNs)
- Payment method (e.g. stolen card used)
There are a lot of scam apps on the app/playstore. Ensure you have the official apps by going to
https://suno.com/Android
https://suno.com/iOS
To fill out the form for investigating the false flags on your account, please see #π¬βgeneral-chat message
Replace section is not working. Possibilty of a refund?
New on here...have a probably dumb question. I had to download part 1 and the extension of a song in order to put them together and fix a few things. Now, I 'm trying to upload the edited track back to Suno to put on playlists, but the upload wants me to trim the song length. Is there any way around this so that I can post the complete song?
https://help.suno.com/en/articles/2409601
Be sure to check out #1294418290348921007 for more tips (and maybe even https://suno.wiki though it is a bit outdated
Do you wish your songs had a different ending? You can make songs longer by using the Extend feature. Best of all, this can extend your maximum song l...
This article is somewhat helpful, but is more about extending the song. But, a question about that. I keep seeing references to "get whole song." But, no matter where I click I don't see that option.
Oh...found it under edit. I was expectiing it to be under download.
a strange quirk with v4, if you have "female vocals" with a genre of music dominated by male vocals like "dark rock", it'll sometimes mix male and female vocals in well done duos. other times it'll ignore the female promt or only use femal vocals in adlibs. i really wish there was a better way to control the ways and rates at which these duo mixes happen
You can increases the chances of it by including gender or names in metatags.
Here is my test playlist (https://suno.com/playlist/11ff3827-bdcf-4ef5-bd04-da8647c6cdb6), most being the different methods (#πβsuno-school message) I try to prompt the genders for duets.
You can also try seeing the tags used by other duet songs (likely listed at https://suno.com/style/duet)
But it doesn't work all the time. In my experience, when it does work (~60% rate), half the time the AI swaps the gender roles
screams
It's not currently an option. The Android version is very basic right now.
You can do it with the browser version, but it's likely to be a minute :cough: before the full feature set makes its way to the Android version
I just want to test the new V4 on the Android
So I guess I'll give it up for while
Hope u guys can make it better on apps:D
A little bit disappointed with android
But that isnt your guys fault lol
So just keep it up and I'll be waiting for new update on Android or IOS
I searched back through the recent couple of days a day and saw many comment from individual solid experienced members who are noting making several consecutive outputs without "shimmer". That sounds more promising and a forward momentum than a worry.
Many new members are needing help and confused regarding how to create songs, issues of using fake apps and billing questions. The remaining are notably having issues which are frustrating from an "out of the box" plug and play expectation.
So it's more a moderate and reasonable forward momentum at minimal loss.
I have still been using the site and getting results which surpass the time/cost needed to get them, despite the work arounds.
And there remains a free level where a few outputs can be tested by anyone daily to check progress at no cost and even rewarded if you get an output you like.
Works for me so far. Keep up the pace.
I guess due some backstage suno changes Less shimmers today, but.. . It comes at the cost of instrumental quality. It went back to v. 3.5. And in second part of the song It gradually gets worse, the voice merges with the increasing frequency of the instrumental sound, and as a result we have a high-frequency voice and instrumental that absolutely does not coincide with what was at the beginning of the song. At least that's how it is for me.
I have been trying to generate songs with v4, but 9/10 songs have this strange, rhythmic, metallic sounds that echoes, mostly after loud choruses. You can hear it here at 03:00. https://suno.com/song/5e7b88cb-b08f-42c7-b0ff-b4bc87ade295
The overall composition and sound of v4 is definitely refreshing and better, but almost every song is then spoiled by this sound
Orchestral build, emotive, soaring melody, heartfelt, uplifting, soft verses, powerful chorus, rich harmonies, reflective bridge, dynamic shifts, tender yet anthemic, spiritual undertone song. Listen and make your own with Suno.
It's a known issue affecting everyone. They're aware and are working on it. It's probably the most frequently raised issue on the discord right now.
"The Shimmer" 
Thank you. Hopefully they'll find a solution soon
Til then... make some tracks about it.
https://suno.com/song/cab2020d-56b8-4489-adf0-a9cc7ad6d96b
Show tune, musical theatre, Disney soundtrack, waltz-tempo ballad, whimsical, nostalgic, uplifting, heartfelt, reflective song. Listen and make your own with Suno.
What does the βthis is a fake appβ watermark mean in songs? What the hell is this shit? I bought a premium subscription, I solve the captcha correctly, but suno generates some shit instead of songs.
It means that your account has been flagged as potentially having accessed their api without authorization. It could have been through one of the numerous other apps or websites that are pretending to be music ai services while actually using Suno on the back end, or just a mistake on their part.
If you're using suno.com directly or one of the official apps and getting those results, they've set up a form for you to report it and (hopefully) get the flag removed from your account... But I'm apparently too caffeine-deprived to find the link. heh. I'll find it
here it is: https://sunomusic.typeform.com/to/zGsJSzsf
Turn data collection into an experience with Typeform. Create beautiful online forms, surveys, quizzes, and so much more. Try it for FREE.
thank you, sweetie. I love you. You are the best of all the people on earth. I'm enchanted by you. You're charming.
I also like cheese?
Ρhineseπ
Sunonians please help! I need a trick to force Suno stopping making melody lines (piano a/o lead synths) in ambient electro tracks. I tried a lot of keywords in the "exclude styles" box, still it always wants to add a melody line. Any ideas? thanks in advance!
"some of you"? It's in all my v4 generations, is there really anyone who doesn't have it?
edit out with Replace Section afterwards?
If you stay away from "loud" genres, like Rock and Metal, and avoid the cymbal-heavy subgenres, the chances of shimmering is relatively low.
I think when I was doing the demos of v4.0 alpha, I didn't run across many of them (though I didn't know of that issue at the time either): #π£οΈβchill-chat message
Yeah, I mean, it's on every cymbal, and it's hard to find genre without cymbals at all? I've tried pop stuff and it's just the same there. So far I'm just stemming out vocals from remastered v4 and layering them with original v3.5 tracks. Works nice btw
It seems to be there on the very first track here: #π¬βgeneral-chat message
although it might be confirmation bias at that point
Nah, that definitely has it but it is also metal. My point was that the chance of shimmering is lower outside of those genres (Rock, Metal, certain EDM/Dance/Trance)
I think adding a prefix to change to a certain type of subgenre may be able to reduce it (see #πβsuno-school message)
For a list of subgenres, I like using https://rateyourmusic.com for their hierarchy
Yes. It often depends, I presume from my experience and comments, on the extent of instrumentation and length of lyrics in your song or genre-- that you simply don't come across issues within a full output.
K, thankx. I don't think v4 does justice to heavy genres with or without shimmer though. Everything's too polished and too plastic, they've Justin Timberlake'd everything
(if not Justin Bieber'd)
Bluegrass doesn't always use percussion.
4 string quartets
A cappella
right
Did you try repeating the instrument in the style prompt with commas? It should give it more weight thus more likely to appear
so a made a song and after the song ended the ai generated it own song in the same track. it was strange and i couldnt work out what it was saying other than "I am funsaco" or some such.
that happens. I got this generated the other day: "There's still your mercy over your ass, you pervert creatures, for your music."
It was a very dark grindcore track too, but I never requested those lyrics
skynet
Are you talking only a melody you don't want.... but an arpeggio is OK?
not sure what you're asking. I was only referring to the shimmer initially
Oh shoot. Sorry. Responded to wrong person. My bad.
Are you talking only a melody you don't want.... but an arpeggio is OK?
I like to coax Suno into generating entire tracks with those Gibberish Sunoese vocals, write english lyrics with the same flow and then do an english Cover of the gibberish track. 'tis fun!
I am not familiar with tb303 but I assume you want this right? Or at least the sound at the start?
https://suno.com/song/aa2289d4-61b2-4e96-8bf3-b9b28c151511
https://suno.com/song/2a1d7fee-4b48-4d8b-a465-c666292c24d3
V3.0 tends to be great at bringing genres and instruments that the later models may have trouble calling.
Here are the attempts with 4.0b:
https://suno.com/song/60561bd9-cb7b-4373-b699-92c4205efbe2
https://suno.com/song/84da4fcc-f309-47bd-99ff-4c94b8a0fb34
it's actually a cool creativity hack
Why is it those babble-scat gibberish songs have also better arrangement and melodies. I've focused on writing lyric arrangements to more emulate sunos gibberish style to try to get a better response.
But if you ask Suno to write lyrics, it's just 4 bars, 4 lines, 8 syllable stanzas. Odd at how it let's go and writes hit (babble) songs if you let it go wild with just a genre.
Yea, I've noticed that
Probably due to it following the flow of the music rather than trying to fit it to the lyrics, I guess?
Yesterday, I ran off 10 acoustic back-to-back gens of babbled acoustic folk songs with my persona , Janis.
enjoyed listening on a playlist style during the day. Easy peasy . Here's the shortest. As always. No shimmer.
Are there any news about the remastering tool that "kills the soul" of our songs?
that one 404's
update worked
very nice!
I'm sad. lol
I got this error yesterday too, when trying to get woodpecker sounds and put "ratatat" in the style
I just wanted to see if Suno'd use that for the cover art gen
if its just a part, thats a somehow usable solution. But Suno depends to make a melodic instrument all the time over the whole song (instead of vox in the instrumentals assume). I wish to build tracks on suno with just the foundation, pads, bass, drums, to develop later on DAW.
I see. Yeah, I don't think that's easy because Suno wasn't really trained on "accompanement-only" tracks. You could probably do it but with lots and lots of tries and (mostly) fails
One hack might be to add some (dummy) vocals, because then it's less likely to add other melody instruments, and vocals are the easiest to separate afterwards
i would like to make tracks with Suno that only are foundations, bass, pads, drums, to work on DAW later with it. No leading instrument wanted. It always brings some up. Excluding does not work either
aaahh, ok, thats a good idea! Thx! iΒ΄ll try it with completely washed out echoing vocals or. something like that! TY!
Is there a similar song in the real world that has those "foundation" feel qualities, that we can try to deconstruct and emulate? Otherwise, you're asking to create a backing instrumental, which Suno may not relate to.
yes, i tried this already. some kind of uk garage, chillstep style, lots of artists there to "emulate". Anyway, I think you correctly analyzed the problem : "create a backing instrumental, which Suno may not relate to." I have to wait to v7.5 π
Somehow this makes Suno not really compatible as a production tool, sadly...
Not until songs can be stemmed cleanly. Right.
but electro just asks for arpeggios, so suno provides. Lol.
the synth arpeggios are ok, i can work with that. the leads i dont want. i may try to go other way round and import fragments of own stuff to see what suno does. in the meantime TY!
Ok. Git it. Here ya go... lol.... yea... guitar lead still comes in at 1:20 and not electro. who knows how. https://suno.com/song/3559d16e-3d70-4be2-888a-9b681a2127b0
Oh. Arpeggios ok. Let see..
yeah i mean, better not, but if...well then okayyyyyy. π
hahaahaaa, the nicest dark ambient intro goes to weird rock
But... guitar!! 
https://suno.com/song/71270621-ba25-430e-b8a4-83c654264b18
https://suno.com/song/7255aae5-7627-40d3-8bdb-866df97b9675
So these are with arpeggios and more new age and slow. So the idea/foundation is there to lay a melody track over. And few instrument of different frequencies which may be stemmed out. Just prompt in that direction can help, I guess. Ok. Good luck with it. There's an answer somewhere in all this, but it might take 3000 credits to find it. Lol
Hey. Its possible have one verse with man and one verse with woman vocal? If yes, how? Thank you
I recommend to use the search. It's tricky. One possibility would be: (from Lumine) It's a topic that comes up regularly. π
You can also put in names or genders if you want vocals from someone else. A few examples:
[Verse 1 - Male] [Verse 1] [Female] [Chorus] [Mrs. Claus] MR. CLAUS: <...lyrics...>Can help indicate to the AI what to sing (though I had like a 60% success rate for duets? And half the time it mixed up the gender when it does work).
tyrannosaur smokin' a blunt?
Suno did give me one... almost π
Thank you π
The type of genre helps as well. Even if heavy metal or other, I can throw in "broadway" or "accompaniment" or such at the end of style prompts to influence a duet. Then tag in the lyrics as well.
Somebody said, Barbershop helps as well. But I haven't tried it. π
I think I started that rumor in June. It's spread now. Lol. But seriously, yes, it opens up both "separate voices" and harmonies, but it's has to be strategically placed towards the end of prompts, to start, then adjusted to taste. Bollywood can sometimes help as there are more % of duet /love songs in that genre.
There are numerous ways to do it, all of which make use of more advanced tags. The styles can have a pretty big impact on how well they work... as can the alignment of the moon and stars. Here's one way: https://suno.com/song/547b778f-4662-435a-9eb1-1f8aed34888c
Spoken Word, Audiobook, Discord Voice Chat song. Listen and make your own with Suno.
"....Dont work at all..." meaning you click and nothing happens.... or the output doesnt return the voice expected... or something else.
Making them or using them? If using them, which way are you trying. 1) Creating a song from the Persona library? Or Clicking create and choosing a persona?
I've forgotten which way works best.. Persona first or lyrics first. Hold I'll test a couple. Brb
To get it out of the way, you can't add persona's to uploaded audio or covers. So any of those you shouldn't have persona as a choice anyway.
Ok. So I took a folk song, clicked reuse prompt, and chose a new persona (Frank Big Band) . It worked fine -- sang the same lyrics.
So are you changing the entire lyrics, or just making some minor changes to same lyrics, a line here, a word there... thn running persona (made from the old song with old lyrics) on this revised version?
Quick check: are the prompts set to the new sound/genre that reflects the lyrics? If that's the intent. Keep in mind, if you add in the steps, persona, AFTER adding lyrics and new style prompts, persona can throw its prompts over the present prompts and sing it's old song because if it.... > i think.
But it sounds like you're revising an old song that you released. So ,
- you made a persona voice from theoriginal (released) song,
- clicked reuse prompt to start a new song,
- clicked create
- reviewed or made changes to lyrics to revise
- scrolled down and added your persona (did the prompts change at that point?)
- click create
.... and you got same old song?
So, I don't know if anyone is interested, but, I started a radio station that is strictly AI music. It's called "The Blender" and if you have a song you want on there, hit me up. I'll gladly add it. However, radio friendly. It does get broadcast on article 15 to a roughly half mile of territory which hits a couple thousand people. And soon, there will be listing on Castbox and a bunch of others including Iheart and Tunein.
And by chance, maybe you're just editing lyrics (by clicking "edit lyrics" ) within the same old song? That would just visually revise lyrics without making any internal changes to any outputs (for your purpose).
Sorry that was for General. My bad.
Personas will not sing different lyrics than it originally had if the changes are minor
Is it just me or is there less shimmer in generations today? Just extended from an audio upload demo I had and it sounds absolutely amazing.
I have noticed it too, not on everything, but it's doing better now
is this the definition of insanity?
is that page 1 of 20? π
Uh... I started adding timestamps but I have given up now
@graceful condor. OK. Here what I did to repeat your issue. All under v3.5
-
I took an old song "So close to water" (Country) and clicked reuse prompts. Added Frank Big Band persona to get a Big Band version. Worked fine.
-
Took that control version. Made a new persona "Big Band Male".
-
went to persona library, clicked on new persona, clicked "Create with Persona" .
-
clicked Create bringing up old song and lyrics with new persona already listed. I DELETED old lyrics and type in two phrases, "This is a test".
-
clicked Create.
Output: it sang Both, this is a test, then added the old song lyrics in full after that. That was your issue you were having.
Test two:
-
I FIRST clicked "CLEAR" in the Create window to start a fresh windows.
-
ran steps 3,4,5 above. Typing in new
lines. -
click Create.
Output: sang the new lines only.... about 19 and 24 seconds worth, and nothing else. Yay!
Ran same test again with another song as in the first round without clearing... I got the two new lyric lines "this is a test be happy" and then 3 minutes of a gibberish song. Both gens.
Ran again but CLEARED the lyric window beforehand.
Ran the steps... it sang only the new lyrics phrase "this is a chance..... journey" and stopped.... 16 and 24.. seconds each.
So, from all that. Start by clicking CLEAR on your CREATE window and go through whatever steps you want (but try mine as well) see what results you get.
using tags be like
just replace 60 with 10
This is my currently unreleased song where I've just been deliberately shoving it in various directions heh
I wanted to name them so I could keep track of what I've downloaded
as a baseline
I'm trying to track somebody down to discuss songs which have a copryright. We own the rights to the songs but Suno hinders our ability to take advantage of its platform. I do not see anywhere to message support or a Product executive. Thank you
It might be a disadvantageous time this month for your team, as the emails are burning up and quit busy at the moment. The main two emails listed in the TOS and FAQs are support@suno.com and billing@suno.com . As your's is a unique issue from most, you never know, you might get a quicker response.
May I ask, as just a curious member, how you are being hindered? Does it recognize you're songs as being copyrighted and not letting you use the lyrics?
I have use some of my older songs filed with copyrights and have had no issues using lyrics on Suno. But then again, my personal songs, though streaming for years, aren't well known.
completely normal

Thanks Dasvon, I'll try those emails to see if you are correct. We are working with well known artist so it is somewhat unique in that sense. But you are correct, sometimes when we use just lyrics it is fine. Othertimes it blocks us. Thus, why I'm here. It is not only a hinderance, annoyance and wasting our credits (less worrisome) but we are not able to maximize the potential of the platform. I've also tried a couple people through LinkedIn. Will see π
You sir, are a god
Great. I've enjoyed using Suno to pitch songs to a couple of start up Jpop groups, but without stems, it would take more or less simply having Sunos version be the demo and frame for creating the song from scratch on DAW/ in studio. But definitely quickens things up a bit. Good luck.
Thanks again. Yeah, we are far past the shopping/etc. Message me if we can help, too
You're very kind. Thank you.
It may be located differently on your layout
Heyyyy! Struck gold. Did everyone know this already. Using personas, will only sing the duration of the length of your lyrics?
I've been trying to make jingles and this seems like a promise. 98% so far.
https://suno.com/song/905bf450-bf73-4092-a7c4-0fe6afa01104
Love you too , brah. Glad it worked.
Yes. That was NOT clearing first...... But the point was, once clearing before you go through the steps.... you should not have to fill in, just to fill in.... it should hopefully just sing what you have and maybe stop at the end.... see my post about jingles..... just above.
But if something is working, you're doing something better.
Dang. Frustrating. At least we have the CLEAR button now which seems to affect the outcome when used beforehand. Again, I Cleared the create window, then went to persona Library first then back to create and entered lyrics.... the prompts were last to review for revising or leaving blank or letting persona fill in.
Ahhhh...it worked then. The solution is a bummer. But that clear button has saved me from many frustrations. Not just persona issues.
So, are ya'll able to see the move to trash option on the browser?
You mean where it's not there.. hahah... yea
Right not in trash. In library
I almost deleted a file permanently. I've reported in bug-browser
I found a work around but it is a pain in the ass.
Show off . π
You made a persona off that song, didn'cha?
Either it's not there like @worn flicker , or I get the restore/delete permanently choices for just Library files.
I'll check my tester account, but that's not the case for songs in my library, unless I made personas off them.
We already put it in bugs
I can't seem to replicate it on my free or paid account, as the ability to "move to trash" is available on every song that I've tested across multiple browsers π€·ββοΈ
It'll probably be there if I turn my phone sideways... haha
The workaround:
Either play the song and delete it from the player or:
Click the menu on the song when it doesn't show up,
Then click off it,
Then highlight the song,
Click the menu again,
It will be there.
Nope still the same...
Now make the menu go away and highlight the song Then bring it up again
It is probably android browser
@charred merlin Hey. Did you see. A way to make jingles! ^^^ or has suno just improved?
I see your using Android
I'll check if I get the same issue on android -- set up the sdk so I can emulate one π
Yes. But there's one thing to have it not show... that's layout formay maybe... but to have Delete Perm/ Retore show.... that's like Mays layout before.
But I'm lazy right now.. later I'll check the laptop
Gotcha
I hadn't tested "motto" or how to force short generations on v4, though I'd noticed that a lot of the old methods from v3x weren't working anymore -- though v4 seems a lot "smarter" in terms of understanding basic requests when you don't throw everything & the kitchen sink at it
Same...lol
When I use persona's with clearing the create field first, it stops after lyrics, so far 100% on 6 gens. I tested personas the other way and no, it just keeps singing. More testing needed but I'm going to enjoy this.
Interest piqued.
ftr -- the "move to trash" missing seems to be just on android browser
Now click off and highlight the same song and bring the menu back up
All is normal on Safari
I honestly haven't tried anything other than chrome. It may just be chrome
https://suno.com/song/19584489-a5aa-45da-907e-7c1b2a963c79
And... well.. not a bad jingle on the fly... hahah... used to write those for a radio station locally when I was.. wow... 20s?
On android
is this "affect the outcome when used beforehand" a common effect when making covers too? I think Suno does not start new from scratch when making a cover several times in a row. It somehow seems to start from the last version made before. Can anybody prove this is right what I think it does?
Dude that is awesome
Ok so you have to use persona to make that happen? I can read above if you explained it clearly
This has potential for radio adverts. Knock 30 out in an hour. We'll make millions! Thousands? A cup of coffee?
With inflation it needs to be millions since millions is thousands these days. Cough...60k vehicles and what-not
Well, I'm wondering if suno just got better.. (or worse and removed hallucination potential) or if I found a hack.
Let me see if I can pull it off without all the crazy, but you are at 100% right, so that will be hard to beat
I joke earlier, but it would seriously be a game changer, or at least less work... So if any radio/marketing execs are reading this..... π
98% ... one gave me gibberish instead of the jingle then said the tagline... but still stopped.
Oh it really would be. I made a commercial for my Marketing course this semester and my instructor loved it...lol. it was all about the jingle that made him so excited.
I'm 43 and back in school. Don't stop school kids.
Christmas, Christmas, ,Christmas, irish drinking song song. Listen and make your own with Suno.
Hey community - I am having a problem getting Suno to pronounce the English word, CULPABLE. Has anyone tried this word in their songs? I've tried to trick out the audio by spelling it LIKE IT sounds and I can't even get it do that... anyone have a work around?
You may be selling "Christmas Levers"
Culp-uhble?
Thanks let me try that.
Hey that is some decent phonetics...haha
I run into it all the time when expanding consonant sounding words like air. Ehhhrrrr works ok.
Something funny. I dont know how universal this is, but my brother is only 4 years older than me. When he went through elementary school they didn't teach phonics. My wife is same. I learned phonics. Talk about a huge deal. It would be crazy to have not learned phonics.
Wow - that word gets destroyed in that song... lol
I still feel bad for people trying to use that word.
I can crop off the song and just keep the intro. Then it will seem clever. π
Ask them to say "Norbert is the culpable squirrel in this caper"
I am at 0% on v4 and v3.5. Nothing under a minute. I know extend cuts off many times, but that isn't what we are doing.
All are around a minute
Sheesh....Louise! What's going on... I was 7.5 of 8. OK. The real test is to see if suno just sings to the end of lyrics and stops, or if short lyrics are the thing... or I have a magic phone or what.
You found something that will probably be fixed sadly.
But at Suno speed you probably have around 3 months of jingle milking...lol
This was an early test... so really 9.8 out of 10. But like on this, a slight intro, and slight ending... like it fits, not just cuts off.
That is cool. I can see where things can get crazy considering that the same amount of credits are used. They would probably err on the side of the longer generation than to get complaints on refunding credits.
Hello! can I somehow extend using another Persona? simulating multiple singers?
You mean like Broadway Style of full on two sets of lyrics running simultaneously?
no, more than a song created using "Jhon" persona, and then extend that song using "Ashley" persona, who would sing the new part.
Oh OK, Duet. Yes, extend is how to do it. https://discord.com/channels/1069381916492562582/1314766265897717780
Those tags how to do it. You change the style on the extend to say either vocal and use tags to get it to sing. It may take a few tries but doing it in sections is the best way to do it.
oka, gonna check it, thank you!
Sadly @polar spade - I still can't get it ... I thought I could use phonix... maybe I should go for PHONK-ix. lol
Hi guys is there someone here from Suno team? I had my subscription for 10.000 credits that said it will expire on Dec 7th. I had 5000 credits left on it. And I had planned to be making use of them max till the end of Dec 7th? and suddenly the day is not over - it's 5:46 pm and it just went to zero mode - like i need to subscribe again. I feel I should have used my credits till the end of the day. Is there a person I can raise this question with?
@muted summit ? perhaps
Hey @icy kindle I just DMed you for the topic I raised above- if you could please respond, will appreciate
The credits expired at the time (of day) that you purchased the subscription.
But it didn't display on the subscription page what time that is - it says generally the date it is valid till. So i won't remember last month at what time i bought. So I just lost 5000 credits it's a lot.
In what phrase is it being use. If I may ask.
I get that, and it sucks, but that's the way that it's been setup, unfortunately.
What does the receipt in you email say as the date... and the time of reply of the email will be close.
Well it will be very bad if they are doing it like that - such customer service for the top tier paying clients is not good. I bought the highest package. I should be either given clarity as per whta time i have the credits for if it is not till end of the day or i should be given the whole day - now it's a choice how they choose to serve the people that bring money to their platform. Loosing 5000 credits is a lot
Yes but i didn't check the validitaty date by my email. I just went on the subscibe page and it says valid till Dec 7. What's my fault? bad interface design.
It should look this in email. What does yours say as your time.
Thanks. I'll try a different way to get your email time it was delivered.
In just the email heading.
The [ To: ] section.... you may have to click an arrow to get the drop down showing details of the time the email was sent.
I am using Gmail, so each may be different.
I understand, well in my case it is Date: Thu, Nov 7, 2024 at 5:25β―PM - but I checked nowhere in the email nor in their invoice it is mentioned that it is until this time but not till the end of the day. they should mention that
Have a look how it is supposed to be done- example- KLING AI
Fair point. I wouldn't know past that. So it's a question for billing for sure . If they charged from Midnight (Nov 5th, 12:01) to Midnight then you would lose half of November 7th and go to, not through,
December 7th (Midnight Dec 6th 11:59).
Equally, you wouldnt gain an extra 6 hours to go through to midnight of December 7th, If they start from 5:25pm Nov 7th. So its there, an exact 30 days by the hour (not date) , but it's likely overlooked by most of us. I wouldn't have thought so either, until I looked at it. Nice if all companies put to the minute on their invoices. Or like water and power bills, credit cards... go to midnight.
Always check will billing though. It's a guess to me.
Well, the easiest is to design it into the website, I wrote to support@suno.com
I know my credits are over on KLing at this time and date and i work my way towards that
now i paid 30 bucks and lost half of it, not bad
Excellent. Hope they give way a little to help you out.
Any way to combat short generations? Lately I've been getting 38s to 1m30s results and that makes me a sad panda.
Do you by any chance have the word SUNO displayed on the artwork generated along with these songs?
@worn flicker hey maybe I'm running a fake app on my Android to get jingle ready gens. Lol...... inside joke from further up chat.
Scrolled through the last 20 and none of them has it.
Hmm. Share one of the short ones so we can see if there's something prompt-related going on?
is it just me or is the shimmer less of a n issue now?
I'm doing instrumental extensions on non-shareable material, unfortunately. This one was V4, but I keep getting it on v3.5 and without artist/song names as well..
Android app or android website through phone not working, i've had to switch to pc
is there any support from suno here?
i just subscribed using the android app, but i haven't received my credits
So, I wonder when people will start to notice the new issue with Suno, I will stay quiet for now
And see if anyone else notices it first
While I agree with the work-around to fix the artifacting, it's sadly done something else, which is more noticeable for my ears, but it's a minor thing, which can get fixed in post
Oh, the almost impossible word to get right is CULPABLE... this is my best effort after 12 "failed tries"... listen-IN around the end of [Chorus 2] @ 03:14... the word comes up shortly there-after in the first line of the [OUTRO].
If anyone listening to this song wants to know what the origin of this song was about, it's a tribute song to a friend whom was claimed by Suicide and thought it was the only way out. Sad-times indeed.
SPELLED AS:
Lyrics from the song...
But have you...
Have you...[Outro]
Your parents ever "kull-pable" deniers, they never understood,
But now I see it all too clear, where you always stood.
Folk Art, Funkel Gart and Saul Pieman vocals, songwriter lyrics. Acoustic guitar shoegaze. song. Listen and make your own with Suno.
How to get the captions of the songs... like srt file. I tried to generate auto caption in capcut, it doesn't recognize the words. For english songs
The video it generates, is vertical and has made with Suno on it.
I have to take back my complaints
I didn't get my lost songs, but they restored my credits and I was able to start producing again.
It took so long to get it dealt with but over the holiday it's not surprising. A rollercoaster.
But now I'm back cranking out tracks that no one but me will listen to probably lol
Has anyone else not been able to generate songs recently? I've bought the Pro subscription thanks to the black friday sale and since then I haven't been able to generate a single song. It's weird because I can still remaster the songs just fine, but when generating it just stays on a loop forever. Any idea where I can talk about this?
Hi ! Have you trying CTRL + F5 ! Sometime it's working !
What is the best way (if possible) to add an epic multiple voiced choir?
just to clarify, you only got your credits restored but not the songs?
Yes
But I keep my lyrics, so I was able to regenerate.... versions of the old lost songs
Not the OG stuff but close enough
Versions I was happy with, that's what matters, I think.
well, at least they worked with you on that
Wait, I just solved it and itβs stupid. I changed from Edge to Chrome
TWO WEEKS and just now the thought occurred to me
you mean you kept the lyrics somewhere in a word doc? that's how you were able to retrieve them?
Yes, I write in a doc, and port my lyrics over, I never write IN suno
Is that the "clean cache" trick?
that's the way to do it, maybe find a way to back up your songs as a wav from now
Oh, I downloaded them all
I literally hadn't got around to downloading the songs I wanted to save when the disaster struck
It was 3 particular tracks, since replicated in a new form.
Yes, that's it! Sometimes the cache is full, and so it lags.
Refreshing the page can unlock a function.
If your problem is something else, then contact Suno support in "More from Suno"
I see, glad to hear they restored your credits at least, that's fair.
Yeah, I had to repurchase the pro license to make sure I own my stuff
But it was the 10 dollar sub version
Made this last night, gonna shill it lol https://suno.com/song/2756deb1-3990-41e4-a62b-a2f1f9daa29d
lofi, synth, vaporwave, chill song. Listen and make your own with Suno.
If I remaster something what exactly does it do for the track? Does it fix timing issues or just adjust the equalization?
Replace Section - Make Same Length as Selection is just plain broken right now
wait you can replace a section and fix stuff?!!!
WUT
Well obviously not when the bug persists but...
I'm going to destroy my credit balance futzing with this.
Until they fix the "make same" I'd skip it .. dont' want to replace a 25 second section with 2 minutes
Just because Suno doesn't understand how to pronounce Tarot
hahahah
Just to end this, yes, I think it was the cache, cause I switched to Chrome, which I barely use, and everything went without a hitch there
Thanks anyway!
i'd like the tokens for my last 7 generations back because "female vocals" is being outright ignored
good luck with that. all your proompting in the world is just suggestions not requirements to the model
Who can tell me what tags can be used to improve and separate parts and where to find the hot keys? For the free version. 3,5?
Please let me know if I'm asking this in the wrong channel, and I'll move it somewhere more appropriate. π«Ά
In the Suno help pages it says:
"In the US, copyright laws protect material created by a human. Music made 100% with AI would not qualify for copyright protection because a human did not write the lyrics or the music."
and it also says:
"If you were subscribed with a Pro or Premier plan when the song was created, you are the owner of the song. You also retain the rights to commercial use for the song, even if you end your subscription."
https://help.suno.com/en/categories/550145-rights-ownership
I recently purchased a year of Suno Premier, and since I'm starting to get a bit more serious about my use of Suno I have some questions related to licensing:
-Is there any issue with publishing 100% AI generated content to Youtube under the (Creative Commons - Attribution) license?
-If I'm not supposed to release 100% AI generated content, what about 99%? Or where exactly is the line at which point there's enough human involvement to not have to worry about any potential licensing issues?
-I'd guess merely prompting with Suno doesn't count towards any human involvement, except maybe if prompting using human-written lyrics?
-I also have paid for RipXDAW as my primary DAW, which I can use for post-processing of Suno generations.
So I'm curious how much I'd need to change things up?
Would a mere pitch shift or change in tempo be enough?
What if all I do is remove a single instrument? (Am guessing not, since this would still technically leave me with 100% AI generated content.)
Or do I need to do something more substantial such as adding in an extra instrument or manually tweaking some notes or something?
Also on an unrelated note, I'm curious if having a year of Suno Premier will my unused credits carry over each month as long as my subscription is active?
Or will unused credits get burned each month?
there is "rights & ownership" channel for these questions it's under "docs, guides & resources" (almost all the way to the bottom) tab I think its best to post there, your questions are very valid, you can start a convo there for sure
ok thanks a bunch, I'll move it in just a moment π
That's a resource section, not open for chatting -- but all of the relevant information is listed there, and you can message the guides or mods if you have questions about it π
I had to brows channels to add since it wasn't visible to me. Thanks for bringing this to my attention so I could add the missing channels.
Unfortunately I don't have posting permission for #1294423389515284552
Hmm, who or where should I ask my licensing questions (posted above) if not here?
Would it perhaps be better to send an email?
Maybe to support@suno.com or somewhere?
I've browsed through the support pages, but don't see an answer for my specific questions, so am a bit confused and mostly trying to cover my ass.
This is where you'd ask -- you can always email support@suno.com, but they've assigned a couple users as "guides" on discord to help explain things -- they're very helpful.
Ok thanks, in that case suppose I'll leave my post here for now.
π new guy here.
I canβt seem to replace sections of a song. I donβt see the option the help article mentions. For context Iβm using an iPhone, Iβve tried using the Suno app, as well the website on Safari, Chrome, and Edge and I still canβt find the feature. Is there a version of Suno Iβm unaware of? Like a PC version? Any help would be greatly appreciated.
There's really 2 main reasons why I'm asking about this:
-
I'd like to be able to publish Suno generations and/or Suno/Human collabs via RipXDAW to my Youtube under the Creative Commons license so that anyone can use the music in their own creative works if they wish.
-
I posted an advert looking to get hired for my first time as a composer working in the indie game dev scene, and quickly started getting bashed by some people who clearly have a bone to pick with AI, though they also raised some decent points. Namely the fact that apparently:
"In the US, copyright laws protect material created by a human. Music made 100% with AI would not qualify for copyright protection because a human did not write the lyrics or the music."
So in this case I need to understand the licensing terms a bit better, so that I'm able to hopefully mitigate any potential risks as far as licensing the music for use in games.
And honestly it seems like the Suno documentation as currently written, is a bit vague for my particular use cases. π€
Persona Generating lengthy Intros like 00:30 π
Hopefully you get your answers here one way or the other but if not, check out "top music attorney" channel on YT
Thanks for the recommendation, I'll check em out.
Is this the right channel? π
Can't really afford a lawyer, but maybe can at least look around Reddit for some kind of "AskaLawyer" or "LegalAdvice" subReddit or something.
Seems better than nothing.
I recently purchased a year of Suno Premier, and since I'm starting to get a bit more serious about my use of Suno I have some questions related to licensing:
-Is there any issue with publishing 100% AI generated content to Youtube under the (Creative Commons - Attribution) license?
Not for songs generated during YOUR subscription period that YOU made as you own full ownership of that and can do whatever you want with it.
You cannot CC-0 any songs made during the free plan or songs that were extended from a free plan generation (and obviously not.ones made by others).
Though you likely knew this due to #1294423389515284552 - specifically https://help.suno.com/en/articles/2871105
-If I'm not supposed to release 100% AI generated content, what about 99%? Or where exactly is the line at which point there's enough human involvement to not have to worry about any potential licensing issues?
The sufficiency of human involvement is for copyright registration with your local Intellectual Property Office (e.g. US Copyright Office).
For example, any Output made from AI as-is will be illegible for copyright protection as of now but rearrangement, manual remastering, etc. can make it so that this modified work can be protected (but not the original Output).
For more info, look up the case of "Zarya of the Dawn" - a comic book that used AI image generation which sets some precedence for AI output cases.
There are several factors when determining rights or ownership to an extension. First, let's discuss the basics of ownership relating to songs made in...
-I'd guess merely prompting with Suno doesn't count towards any human involvement, except maybe if prompting using human-written lyrics?
Correct
-I also have paid for RipXDAW as my primary DAW, which I can use for post-processing of Suno generations.
So I'm curious how much I'd need to change things up?
Would a mere pitch shift or change in tempo be enough?
What if all I do is remove a single instrument? (Am guessing not, since this would still technically leave me with 100% AI generated content.)
Or do I need to do something more substantial such as adding in an extra instrument or manually tweaking some notes or something?
Again, this would be all for Copyright Registration with your governmental IPO and it is up to them what is sufficient or not.
The more manual work you put in, the more likely that modified work would be accepted for protection - but the original musical Output from Suno would not be.
(When you submit the registration, it will ask you about each component. So your lyrics, manual modified song, etc. Can be protected - see the link earlier or look up about the Zarya of the Dawn case #πβsuno-school message)
https://youtu.be/HlGIxLH1K-M
Useful timestamps:
- 05:10, she is talking about the user giving all (irrevocable) permission/rights for Suno to use the Inputs submitted
- Around 08:10 which covers the Output's ownership is given to the paid members.
- Free tier info was around 10:12
In this episode of the Top Music Attorney Podcast, Entertainment Attorney, Miss Krystle breaks down Suno's Terms of Service, focusing on the rights you give up and retain when using the platform. They explain how uploading content grants Suno broad licensing rights, including the ability to sublicense and distribute your creations. Miss Krystle ...
One question though, the shimmer..... can there be a "shimmer removal tool" for songs that already exist? Some of my best ones have shimmer to h3ll.
Edit tools like Replace and Crop are not designed for mobile view at the moment.
To access these tools on the website, you need to be in Desktop Mode (otherwise you will get redirected)
The best we have right now
#1309224710692929548 message
No, no, and no.. Woodhouse is plain and simple dyno, and for the most part I setup his rules.. π
I've had some good experiences with downloading Stems and merging them back together with Audacity. But that doesn't solve the problem in Suno. Extending them with 3.5 works so and so. ^^
Thanks a bunch for the response!
I've been playing around with Suno off and on since not too long after the public V1, but yea I've not done (nor do I ever intend to do) anything with free generations outside of sharing with some Discord friends.
So am only discussing Suno Premiere generations here.
That said, the differences between Ownership and licensing rights and copyright stuff etc, is a bit confusing to me, but am trying to learn.
-I'm curious what the legal differences are.
If I had to guess the main difference, in the case in which I own my Suno generation but am unable to copyright the track as is, does this mean that if someone else were to somehow get their hands on the song then they''d be able to use the song as they wish due to the lack of copyright protection?
-That aside, if I understand your response correctly, then as long as I'm using Suno Premier then I shouldn't have any issues publishing even fully 100% AI generated music to my Youtube channel under the (Creative Commons - Attribution) license?
I'll give the vid a watch soon, after i finish reading the rules for r/legaladvice. π
Thanks for the recommendation. π
That said, the differences between Ownership and licensing rights and copyright stuff etc, is a bit confusing to me, but am trying to learn.
Yeah, I get you, I was confused with the same concept before.
Ownership means that all the rights associated with the Output (100% Suno's for the free plan) would be yours, as if you created it.
In order to publish your songs on YouTube or when you use Distributors, you need the rights (and/or neccesary Licenses) to be able to distribute it (not to mention the monetary aspect).
For Licenses for Remixing/Cover/Samples, see https://dittomusic.com/en/blog/what-license-do-i-need-for-my-music/
Not that it matters for Suno as Suno TOS requires you to have full rights for your Inputs.
Free plan users can upload to YT so long as they do not monetize the Output and give Attribution (e.g. Made with Suno in the comments) - that is covered in the TOS and a few other places.
Although you do have Copyright for any of your own creations, e.g. your lyrics, you cannot sue anyone over it unless you have proofβand that is mainly what copyright registration is about (and usually is country specific though there are international copyright treaties)
does this mean that if someone else were to somehow get their hands on the song then they''d be able to use the song as they wish due to the lack of copyright protection?
If someone got their hands on your original Output, you cannot sue them for using that Output (either as-is or if they modified from that).
If someone modified your successfully Copyright-protected work, you can sue them (and it would be up to the defense to prove they used the original - which they can't if you never link/Public the original)
-That aside, if I understand your response correctly, then as long as I'm using Suno Premier then I shouldn't have any issues publishing even fully 100% AI generated music to my Youtube channel under the (Creative Commons - Attribution) license?
Correct
I think that SoundCloud has updated their algorithms and they can now understand if the track was made with Suno (as Suno adds audio watermarks between wavelengths) and they're now limiting the reach of AI generated music. Maybe it's time that you stop putting those audio watermarks on songs Suno, especially for Premium users?
it's annoying, especially when I've used the Upload Audio feature, aka given the input by myself, in terms of audio, just by prompts
but I've experimented with SC. When I upload a song made by me in Ableton fully VS a Suno track that sounds so much better than what I can make in Ableton, then the one made by me gets much more organic reach. And to make this test valid, I produced music from the same genre
It's fcking annoying tbh.
You should put that in #π‘βsuggestions-web - not many devs visits this channel but they are more proactive in the suggestions and bugs channels
Though it makes sense SoundCloud is doing this.
They have AI partnerships with Fadr, Soundful, and Voice-Swap β and would like to reduce their competitors creations https://press.soundcloud.com/234082-soundcloud-partners-with-fadr-soundful-and-voice-swap-to-introduce-artist-first-ai-integrations
Soundful, lmao
oh god
most horrible company out there
I used to be their customer, aka subscribed to them
team has no vision, no product updates, no transparency
when asked about new things, they only promise that 'big news soon!' but no news arrive
no information what is happening, no updates, no nothing
I'm just bitter, because I subscribed with them for a year, hoping to see how fast it develops etc
but have barely heard anything new over the past 6 months, when they launched a new tech house template and few other things
I recognize that the Suno docs specifically mention Youtube, that with Pro/Premier we get get commercial use rights for monetizing on sites such as Youtube being specifically mentioned.
However, since it doesn't specify between Youtube's Licensing options, I assume this applies to the Standard Youtube License but am not sure about Youtube's Creative Commons License:
"Creative Commons licenses can only be used on 100% original content."
"You can only mark your uploaded video with a Creative Commons license if it's all content that you can license under the CC BY license."
https://support.google.com/youtube/answer/2797468?hl=en
Changes to video attributions: The view attributions page will be discontinued after September 2021 due to limited usage. You can still add attribution to your videos by updating your video descrip
My reason for concern here, is because it seems like I might need to be able to copyright the music in order to publish under Creative Commons (CC BY) without risking potential problems:
"CC licenses are operative only where copyright, sui generis database rights, or other rights closely related to copyright come into play."
https://wiki.creativecommons.org/wiki/Considerations_for_licensors_and_licensees#Considerations_for_licensors
"Creative Commons licenses can only be used on 100% original content. If there's a Content ID claim on your video, you cannot mark your video with the Creative Commons license."
https://support.google.com/youtube/answer/2797468?hl=en
"Licensors should read and understand the terms and conditions of the license they choose before applying it. Licensors should also secure all rights necessary before applying our licenses so that the public can reuse the material as expected."
https://creativecommons.org/licenses/by/4.0/legalcode.en
"Make sure the material is subject to copyright or similar rights.
CC licenses are operative only where copyright, sui generis database rights, or other rights closely related to copyright come into play."
https://wiki.creativecommons.org/wiki/Considerations_for_licensors_and_licensees#Considerations_for_licensors
"CC licenses are copyright licenses, and depend on the existence of copyright to work. CC licenses are legal tools that creators and other rights holders can use to offer certain usage rights to the public, while reserving other rights."
https://creativecommons.org/faq/#Can_I_use_CC_licenses_to_license_rights_other_than_copyright.3F
Changes to video attributions: The view attributions page will be discontinued after September 2021 due to limited usage. You can still add attribution to your videos by updating your video descrip
"Creative Commons licenses are copyright licenses β plain and simple."
"Without copyright, these tools donβt work."
https://creativecommons.org/2010/06/30/response-to-ascaps-deceptive-claims/
Last week, the American Society of Composers, Authors and Publishers (ASCAP) sent a fundraising letter to its members calling on them to fight βopponentsβ such as Creative Commons, falsely claiming that we work to undermine copyright.* Creative Commons licenses are copyright licenses β plain and simple. Period. CC licenses are legal tools that c...
Shimmer is getting less and less now. songs which shimmered to all hell before are not now
The TLDR form of my main question is:
"Are there any potential issues to be wary of if/when publishing 100% AI generated content produced via Suno Pro/Premier to Youtube using specifically the Creative Commons - Attribution (CC BY) license?"
That said, if we can confidently resolve this question, might be worth updating Suno's documentation to include the answer.
Because currently despite mentioning that we get ownership rights (Whatever all that entails) and commercial rights (Using Pro/Premier) able to publish and monetize content on Youtube; it doesn't specify if this only applies to Youtube's Standard license, or if this also includes publishing to Youtube under their Creative Commons (CC BY) option.
Hey so, I'm trying to generate a song on the android app, but it's silently failing and refunding the song credit. There's no error message, I have credits remaining, and I'm using 3.5. What gives?
That really sounds like a question you should be addressing to YouTube, not to Suno, especially as the question is more for their end and none of us here are lawyers.
As far as Suno is considered, your are granted all rights, titles, interests, etc. to the Output you made during a subscription plan.
Did you try reproducing it on the website to see if it gives you more information? The app is very barebones right now and I would not be surprised if they forgot to include some error messages
I'm currently havin meself a think on the porcelain throne, so no, I haven't tried the website yet. I'll give it a shot in a bit and see what comes up
You think Youtube would respond to such a question?
Doesn't really seem like a Youtube question to me so much as a Creative Commons / Licensing question.
Seems like the question would be better directed at Suno support, the Creative Commons ppl, and/or a legal expert.
Seems like the answer depends on if publishing under (CC BY) requires the licensor to own the copyright for or at least be able to copyright their work. Or not.
And whether or not the ownership rights / commercial use license provided by Suno is sufficient.
Hm.. π€
Could someone please link me directly to, or point me towards where I can find the exact / full License terms offered by Suno?
I may be overlooking it somewhere, but not seeing the full license terms in the Suno docs anywhere.
Did you see the video I linked earlier to Tip Music Attorney? She covered the TOS.
The full TOS can be accessed at https://suno.com/terms - also linked in the "More by Suno ..." link on the website
We are building a future where anyone can make great music. No instrument needed, just imagination. From your mind to music.
No I haven't yet.
Apologies, I've been distracted trying to figure all this stuff out.
Will go watch it now. π
And then will comb through the ToS you linked. Thanks. β€οΈ
And apologies if I'm being annoying or troublesome with my questions.
TBH am a bit paranoid after my indie game composer advert got ganged up on by AI haters. π
You're going to get haters no matter where you go on whatever subject. Gotta let those things wash right over you
I've got a paid account but I can't download WAV - the option is there, but says "download failed. please try again." I noticed MP3 works fine, Anyone else had the same issue?
Seems to be browser related bug, if you can try another browser, may work.
I'll try browser change - tried flushing cookies, cache, etc... But I'll troubleshoot further at browser level. Thank you!
I have a very immature and embarrassing request: Can Suno generate scores?
Already answered in the other two channels you asked on without waiting
Well yea sure to an extent, unless the haters have some kind of point.
IMO is a good idea to not feed the trolls, to not get bothered by mean/hateful comments, but also attempt to always be receptive to negative feedback in order to help correct any errors and/or generally learn and improve things over time. π
UPDATE: NOT A BROWSER ISSUE... I've got a paid account but I can't download WAV - the option is there, but says "download failed. please try again." I noticed MP3 works fine. I have paid Pro account. Worked fine days ago and now none of the tracks would download the WAV format. Anyone else had the same issue?
Do you have "SUNO" image watermarks or audio watermarks saying to use Suno.com in any of your recent works?
no, never seen those
no, never seen those
Did you try downloading through Firefox?
I usually see people complain about that issue on chromium-based browsers (may be related to the filesize of the wav)
Just tried - it worked! THANK YOU! OMG it's been hours of head-scratching. I am forever profusely 1 000 000x grateful.
Just curious, but is there a reason to prefer .wav over .mp3?
wav is uncompressed full sound. mp3 is compressed and smaller file.
wav is better for performing djs and artists. mp3 is better for standard normal people who rather have more storage space in their hard drive and dont really care about the fullll spectrum of sound. so its a quality vs performance preference
Tyvm for the info/clarification. π«Ά
For Some reason it's not letting me download my wav files. I keep getting a time out error. Anyone know whats up with that?
Anyone else having a problem with the "Replace Section" function? Over the past two days, it hasn't been limiting its replacements to the length I selected. And I made sure the button for that was in the proper position. I've spent a bunch of credits apparently because this feature has a glitch or I'm doing something wrong that I have no clue about.
Finished watching the vid you linked and then combing through Suno ToS.
The TLDR answer to my question (||"Are there any potential issues to be wary of if/when publishing 100% AI generated content produced via Suno Pro/Premier to Youtube using specifically the Creative Commons - Attribution (CC BY) license?"||) seems to be:
Yes, there are potential issues to be wary of.
The long answer to my question:
According to https://suno.com/terms and https://help.suno.com/en/articles/2746945
Suno seems pretty clear about how:
"Suno makes no representation or warranty to you that any copyright will vest in any Output."
and:
"Music made 100% with AI would not qualify for copyright protection because a human did not write the lyrics or the music."
||"Subject to your compliance with these Terms of Service , if you are a user who has subscribed to the paid tier of the Service, Suno hereby assigns to you all of its right, title and interest in and to any Output owned by Suno and generated from Submissions made by you through the Service during the term of your paid-tier subscription. However, due to the nature of machine learning, Suno makes no representation or warranty to you that any copyright will vest in any Output." ||
So while I'm not 100% certain that 100% AI generated content can't be published under Creative Commons, but unfortunately this does seem to be the case. =/
Nor am I yet certain how much human involvement is required to enable copyright protection, but I'll continue looking into it.
I suspect the answer to this followup question might vary depending on legal jurisdiction. π€
We are building a future where anyone can make great music. No instrument needed, just imagination. From your mind to music.
Copyrights are complex, and vary by region! To ensure you have the most up-to-date information, we suggest reaching your local copyright office before...
Regardless, this no longer seems like a question for Suno, but instead seems like the sort of thing I should probably ask some kind of legal expert who's knowledgeable about copyright law. π
is there any change with the voice in V4??? its seems very much on high frequency (vocals)
Thanks for all the info to everyone who responded to me here. π«Ά
I'll try to leave you peeps alone now, unless I find out some info that seems worthy of sharing. π
also i saw a new error.. with stereo..
Wtf guys is this HIGH filter on every V4 songs ??????
Guys for me its now near to be impossible to have a clear full sounds.. idk why that fkin high frequency thing has pop ....
plus lyrics that arent in english sound...not so good
Feel like prompting changed from v3, but can still get same sound
So v4 still broke?
I remastered my song in v4 and nothing has changed
I can confirm this. I'm experiencing the same issue and have seen it in many other songs as well. It manifests as a harsh, high-frequency noise and seems to get worse towards the end of the song. It's very painful for the ears π.
In red: Shimmer
This is one example of what I mean.
The βvinylβ sparkling is not what I mean itβs part of the song.
Dreamy Indie Pop, Ethereal Atmosphere, Floating Melodies, Soft Dynamic Build, Intimate Vocals, Atmospheric Production, lo-fi song. Listen and make your own with Suno.
Is there anyway to resolve WAV file down load delay?
after i click download wav it stuck for a while before started download
Suno website throwing an error when attempting an extension on a V4 remastered instrumental that's been cropped
I added two extra tags that weren't present before
It appears to let me extend the non-cropped version
Is this for real? Why is the max page for search 11 when there are more songs? It stops at showing only 200 songs despite there being more.
When trying to go to page 12 (from_index: 220), it returns empty, despite clearly saying the total_hits being 500+ for that search in my account.
It seems search only works up to "from_index: 200", max 200 songs. That's insane, is this a known bug?
i am and some of my friends have problems with it
Just a logistic note: If you would right-click and replay rather than re-post only, it would be easier to follow your issue and chain of responses. π
Too many variables in your specific post to know, frankly. Since it was 3 hours ago, time may have solved it as well. So, you can reply an update if you wish. It's a common error covering a few issues.
Excuse my ignorance, but where would it show the total hits of the 220 visual limit? I've not seen a search tally indicator before.
To verify, my page count goes to 11 as well.
One of my v4 songs was played in a club yesterday by a friend DJ, the mastering of v4 is incredible the sound was crisp and the bass went through everyones bones lol π
I had issues in my modest garage setup with 3.5 especially the levels between voice and synthlead was sometimes unplayable at higher volumes. Whatever issues v4 has, the mastering improved a lot
Have the same sharpining issue getting increasingly worse throughout the song.. Like overly processed percussion and sibs.
Hopefully, this is something the devs are aware of and will address.
Keep downvoting stuff like that.
yeah, hopefully. Hard to sit on my hands and wait though π I think their downvote/bug report log is pretty full though π
It's not a log that gets reviewed by humans, it helps give information to train future updates. The more reports you give, the better data the AI has.
I know π
Given I'm ignorant of most things, how much does AI "fix" or improve, based on the downvoting without any human involvement. I.e. self-correct. None?
It likely doesn't self-train π
less rushed response -- though not impossible, it's unlikely that our reports are immediately impacting the model. https://en.wikipedia.org/wiki/Reinforcement_learning_from_human_feedback
Even train wheels from 1860 are self-correcting. Come on. Lol. π hmmmm but they do go off track and kill hundreds sometimes. Ok. Never mind.
And good morning everyone.
y'all ever run into doing a cover that is just super stubborn and won't deviate from the original much at all?
for me SUNO V4 is no longer fonctionnal due to very harsh high frequencies and getting harsher while the song continue, there is no chance to create a perfect music or at least, a viable one. We need to raise our voice as fast as possible ! It's like they tweek things and never test what it looks like 
I have the opposite issue: it wonβt reproduce the melody in my source track β¦ well not until 72 random tries. Then I feel obliged to accept that one because, who knows when or even if youβll get another match.
hmm
Paul, do you have that Harsh issue too or its me only?
bro v3.5 is really bad when you tasted V4 lol
it was near to perfection and since 3 days, it's a total sh*t
lol one of the track it gen'd was 24 seconds long.
also regarding to all recent changes, it seems V3.5 has downgrade, or maybe i'm too used to V4 idk.
apart from the shimmer issue, I've been getting really tinny results. Like the drum player only knows how to work the cymbals.
I do have it sometimes, but my trick is to put a decade at the beginning of the prompt, like β1990sβ and then whatever. And sometimes that fixes it. Also sometimes I donβt have the shimmer. Tbh, this could be more like a rain dance than an actual method: rain just happens to come when the guy is dancing.
v3.5 hasn't changed. They're focused on addressing v4 issues. The shimmer issue has ruined most of the stuff I've generated. This high-frequency noise isn't good either. V4, being a beta version, is still being worked on and will likely receive further updates before it's fully released.
i see... if you want a little turnaround to get better result with V3.5, go to cover your final song with just Pop,[voice genre] and it should get you a cleaner overall sound
in what genre of music you don't get the shimmer?
Iβm focused on the performance, not the mastering (just my preference), so I am ignoring those shimmer and too much high frequency backfield noise for now, but β¦ just curious, do you find that v4 has less inventive instrumental parts than 3.5, and that the balance between lead vocal and instruments is way too much vocal? Iβm trying this: create the performance in 3.5, then βremasterβ in 4. What I see is that 4 will miss notes here and there, like fail to play the correct note in the melody, but still be a βcorrectβ note for the harmony - sometimes. So, βremasteringβ sometimes means rearranging!
i get a system error when I try to make something with v3 atm, is it a known issue or is it me
the note thing was happing even in cover in V3.5.
for me thoug
Paul maybe its not the place to talk about that, but you're working in music tech AI as product designer?
I noticed that too, but when they call the action βremasterβ I donβt expect rewriting. But Iβm accepting. Itβs an evolving tech. And getting models to be creative is, for me, far far more important than getting good mastering.
true, thing is, i'm french and i'm complaining a lot !!! x)
not a big problem rn V4 is a bit messy
https://suno.com/song/be7fdc21-ff40-40f8-b433-2cec505401ee <Depends on how patient you are
electric grunge gritty song. Listen and make your own with Suno.
Anyone have the same issue with me, only generated 8 sec and credits didn't refund to me.
is there a wat to get the song from 4.0 similar to remaster but from 4.0 to 3.5
no
do you have an example?
Like this. The 8 sec one is totally useless, but they still charged my credits.
a link to it directly would be nice? I assume you're doing instrumentals?
what is the best way to get suno to do a quick end in an extension? i tried abrupt end but i think i got it wrong somehow
well, you could just crop it?
the issue is the tails
you mean liek little insturmental bits at the end?
yes
so.. maybe add "[end]" before it?
hmm
so I have this: for one of mine, and it ends after only a few beats of the final line. if you're doing instrumentals.. I honestly don't know what can be done in that regard.
What's up with the vocals? They sound horrible again.
in v4? it takes a few tries for just about everything to get it right currently
yeah. distorted and just strange
#πβsuno-school message this is what I had to do with ONE song to get to to behave
https://suno.com/song/ebe54331-1459-4b90-8e24-6f272d03f65a <I think it sound pretty good though
grunge, mysterious, haunting, dark, electric, raw, female vocals song. Listen and make your own with Suno.
It does. I have tried remastering and regenerating but still sound like garbage. I see now that more people has the same problem so i guess it's better to wait a few days instead of throwing credits into the void.
oh it's definalty not perfect. but you CAN get it done. that's more my point. If you need the e credits to be held I would keep to the v3.5 myself and remaster later when it's all fixed up.
I am sitting on 120 after all my remasters and a few new songs.
I have 3 more to make, they'll all be 3.5 and I'll remaster them later on
Yesterday I did some experiments in making vocals sound like they've been chopped and sentence-mixed, like in Harder Better Faster Stronger
Iβm having an issue with my Suno Pro Plan subscription. I subscribed through the App Store, and it worked fine for the first week. However, my account now shows only 50 credits and no access to GPT-4, even though the Pro Plan is still active.
Iβve emailed support twice but havenβt heard back yet. Can someone help me resolve this or point me in the right direction?
That's really weird and frustrating. Aside from emailing, I'm not sure what to recommend.
I also had 2 referred friends make more than 10 songs each and the 250 credits hasn't shown up for anyone.
Itβs so incredibly frustrating. Iβm hopeful I may get assistance via Discord.
Sorry to hear about your issue.
Are you using mobile or browser or what?
There's some simple troubleshooting you could try, but I don't want to assume you haven't already.
Like, shift-reload the browser, clear your cache, sign out and back in again (on whatever platform)...
And did you use all 2500 of your credits in that first week?
You got it tried all that and restoring my purchase in the App Store. Reinstalled the app. Nothing. π¦
Hmm.
Turn it off and back on again? ^_^
Sounds like an account problem then, assuming you didn't use all your credits. Keep in mind that once you use them all up, you get the 50 credit daily reset like everyone else. The fact that it isn't letting you do v4 tells me something else is going on, though.
Looks like I'm not the only one who has problems with the quality of the music, especially the vocals.
I'm new to Suno, how long do I have to wait for improvement?
For now not much, the heavy shimmer seems much less now than it was just 40 hours ago, but the new thing in town seems to be overly harsh cymbal / highhats. Stay with 3.5 for better quality overall and re-master to v4 once it improves. OR stay with v4 and test out possible workarounds on your own π
V4 is much better than 3.5 in "quality," shimmer be damned
Bad news for you, one of V4's selling points is improved vocals. Obviously it's still not perfect but unfortunately this is as far as it's come
That said, if you're just listening to your own outputs with a free account, you're not using V4 and not hearing the new improved vocals
On some points i totally agree but at least for my use its still beyond repair, even in RX 11 advanced. due to the high hz overall sound
It does sound more toned on regular speakers and headsets though, but on studio monitors its razors and knives, specially if you go over the 2 min mark
https://suno.com/song/d6080760-1f77-46e0-a011-a5c137d98acc
In this one you can hear the AI vocals having trouble in places but you can also hear the nuance it adds in order to sound like a given genre
emo, pop-punk, angsty, 2003, emotional song. Listen and make your own with Suno.
Shimmer is there but not as sharp, but also a few days old
Oh, you mean the shimmer hurts your ears with the headphones you're using? Interesting.
not the shimmer,
I have a pair of DT770s for my gaming PC, so maybe I'll use those to listen for what you're talking about
But maybe your studio monitors are playing something nasty that most headphones can't
but hey.. i also know that every experience and render is wildly different, we do not all render the same genres or use the same prompts. So you and others might be getting good sounds π Can only speak for my self
Go piss in your own throat and drown in it, Woodhouse
I have heard that rock fares a great deal better
There was the shimmer, now there is this high-pitched noise. I guess training the model is like trying to steer a boat and they're oversteering it a bit and it takes a while for it to actually head in the desired direction.
I've run into a different quality problem that's in both V3 and V4, which is that extends cause muddiness
yeah but its tantalizingly close though π Like threading a needle while jetskiing π
Well let me gen something and see if I hear high pitched earrape instead of shimmer
β Warning logged for catgirlkazu. I couldn't DM them. || The swear filter exists for a reason. You may not like it but you are aware that it exists and responding to it in the way you did is inappropriate. @worn comet
Hi I'm new here, has Suno addressed the shimmer issue at all?
Oh my heavens what the heck did I just hear?
https://suno.com/song/20c19105-b7b1-40c3-a562-01b8ff12202a
https://suno.com/song/344d0694-ae85-49ec-a2a8-faa802880d00
You tell me, do you hear it in these? I think I hear it, but it's more like a quality problem than an artifact from the AI
jungle dnb, drum and bass, EDM song. Listen and make your own with Suno.
jungle, drum and bass, EDM, -melodic song. Listen and make your own with Suno.
Doing a brief skim through these: I don't hear shimmer, but I do hear intense high frequencies, which seems to be the newer issue being reported. Shimmer is more like an annoying intermittent spring-reverb ratatatatatat noise.
Yeah, that's what I was wondering. Hope they change it soon, because this is way harder to listen to than the shimmer
Percussion from v4
HELP! Replace section makes later part of song βskipβ
- detail I have a song that is βnear perfectβ but I had to change the verse lyrics slightly to make them flow better based on how Suno βsingsβ them. The insertion and replacement are perfect, but later in the full song, it βglitchesβ forward skipping a chunk of the chorus.
I know this is a beta, but Iβm wondering if anyone has found a workaround.
Weird. Percussion in my songs do not sound like that, that was awful!
Oh my god. Oh my god you guys.
One click. No trickery. Lyrics exactly as you see. Almost every single gen handles it THIS beautifully.
to be fair is was the percussion stem only and in the late part of the track, some of the more metallic tones are remains from the vocal but the rest is cymbals/high hat which is giving me issues today. My condulences to your ears π
Currently having issues uploading new artwork.
No, still has 95% of my credits available.
Hmm. Well... then I'm out of ideas.
LOL wow I love it
I'm having an issue creating
I wish I had that kind of luck. I blew through all my credits in a week trying to get one song to come out right at all.
Try again? Sometimes errors pop up and then go away when you just click again.
Got the same message
Reload the page?
I already tried that ...
https://suno.com/song/f6d6ee8b-4700-46dd-ab54-7845f3735f7a
It wasn't even a fluke. I got a whole bunch
Jungle, drum and bass, dubstep song. Listen and make your own with Suno.
This is the best one imo
is the website working for ya'll? i've been waiting for 20 minutes and it keeps loading, my wifi has no problems and i've refreshed it over and over againn
Works for me.
anyone having problem like me ofr replace part of song. i want replace 2-3 words but replace change more part automatically and change composition also.
@charred merlin sir, I am using the "replace" function, but more text is being replaced than the selection I made. This is affecting the main composition. I have tried using the "Make same length as selection" option, but the issue persists.
Is there any other option or solution to ensure only the selected text is replaced?
With Replace, I've found you just have to keep trying. It's not very reliable, and insists on replacing 10-second chunks.
Merhaba admin varmΔ± yardΔ±ma ihtiyacΔ±m var
https://suno.com/song/d7ced6cd-5d00-4a25-8549-078652417c39
Behold my masterpiece.
Jungle, drum and bass, hardstyle song. Listen and make your own with Suno.
Help please @cerulean root
?
I subscribed to Premier 1 week ago, the quality of the songs released has deteriorated and when I add my own voice, the sub melody cannot be heard at all, only my voice is present.
? @cerulean root
I dont work for Suno...
Any advice to prevent suno puting a Guitar solo anywhere? im using the "exclude" Guitar solo, guitar etc. But it bypasses it. I just want a tranquil instrumental section on a metal song.
Maybe put a specific instrumental interlude there? Other than guitar. But with Metal it could be difficult to omit. I had some luck with Dark Cello. But that's more Gothic Rock than Metal. π
maybe haunting banjo
...since we can't prompt for the kazoo.
try in exclude style -guitar solo
I see it in [ ] brackets, might not always work as well, that's not "exclude style" , unless I missed him saying it in other messages
Yeah, he did. The message was just a joke out of desperation. π
@outer cosmos Unfortunately, this post from suno-memes shows this is a common issue π
There are many tools in the Suno Workshop, but this isn't one that I personally use often.
I know the team has expressed interest in making it better, so I have hope for the future that they will improve this feature over time πͺ
Dear [Suno Support Team],
I hope this message finds you well. I am reaching out regarding my recent subscription activities with Suno. Over the past 30 days, I have purchased three different plans. Specifically, I had initially subscribed to the monthly plan just five days before deciding to upgrade to the annual plan.
As the annual plan took effect immediately upon purchase, it replaced the monthly subscription and any associated credits that remained unused. Considering this overlap, I kindly request that the unused credits from my previous subscription be added to my account or adjusted accordingly.
I appreciate your attention and consideration in resolving this matter. If you need any further details or clarification, please do not hesitate to reach out.
Thank you for your support!
Best regards,
[Marcio de Campos]
Please e-mail support@suno.com
New problem i feel like drum got really worst with last update cant hear cymbal as clear as it was before
Is there secret tags or something for not diverting from lyrics ect?
blown through 100s of credits and it still will not say the words in verse 3. when do they renew on pro?
we need to know
I made quite a few songs using ReMi, I thoroughly enjoy the chaotic outcome from the crazy lyrics π
By the way, quick question guys; I have an annual subscription. If I buy extra credits on top of my monthly credits, would the monthly ones get used first and then continue with the extra credits or would it be the other way around?
Can confirm this to be the case. Sub credits are used first, then extras.
Extras also do not expire.
https://suno.com/song/4e1509c9-3c16-4011-8512-42e1917e5f74 is it me, or does it sound like the gate of hell opening up at the end of this song right st the 3 min mark
ambient, bossa nova, sultry female voice, acoustic guitar, percussion song. Listen and make your own with Suno.
βIβm reaching out to express my frustration with the lack of support provided by the Suno AI team. Despite being a paying customer, I have encountered issues with the software and have not received any assistance.
βIβve tried contacting support through Discord, Twitter, and the official support channels, but no one has responded. This is very disappointing, especially considering Iβm paying for the service and not using the free version.
βI expect better communication and support for paying users. Please address this issue promptly and improve your support system.
There is no official support team here, this is primarily a community-ran server.
It's currently the weekend so you are not likely going to receive a reply. Typically, it takes a few business days (US Eastern Standard Time, UTC-5) for them to respond in normal circumstances.
My recent email took about 8 business days to respond to, they are currently swamped due to the long weekend before and Black Friday deals iirc
guys, i just subcribe suno via google pay 3 hours ago, but my account still free version. why it happened?
I hope you at Suno will fix the hissing effect after 3 minutes.
A snapshot of the current "issue" with V4 as of today. On friday the shimmer were drastically improved (much less prevalent) , however the tracks rendered from there went from shimmerish to having a high pitch / syb sound in the percussion (and some times the vocals). The general experience is that the track starts off well with balance and sound, and then starts to fall apart around the 1:30 - 2 min mark with an increase in the hz. When splitting the tracks (have done 50 some since friday) the same story is apparent and the issue can be seen. Noticed the increase in brightness in the upper right part. (higher hz/end of track)
V4 are miles ahead of v 3.5 though, and are so so close to "perfection" π
fyi. do mostly pop,soul,jazz, reaggae tracks so have no real idea if its the same in other genres?. π
Is SUNO aware of this issue? Hopefully theyβll fix this ASAP
Spent 200+ credits due to this cuz I thought the issue was my prompts
im sure they are, and since they did something with the regular shimmer on friday im sure they are looking into it, the effect/issue shown came as a result of the fix, likely a change in instrument pack used i guess. Listening back on tracks rendered since the launch of V4 (both my own and others) i seems as the best renders V4 ever made were right at its launch.. then after some time the shimmer (percussion reverb) came along, and now we are at the hiss / syb phase i guess. Honestly liked it best at launch. π
Nah its a Suno thing.. What genre do you make tracks in and male or female vocals?
Iβm making mostly jazz/rnb/neo soul
is there a way to add weight to the prompt ?
Both female and male vocals
Not that i have found. π¦ Only the devs know for sure. I have a sneaking suspicion that its introduced by them using another percussion back for the AI though. I have had luck with hiphop/rap before havnt rendered a track in that genre since last week though. Another thing you could try is to make a short track i.e. 1 min or so and the extend that to get a better outcome in the end of the track. But thats just me brainstorming, havnt tested it
Thanks for sharing π«‘
np.
The only way i've really got to avoid shimmer. Multiple stems from covers, pick the track segment with the least amount. Im sure this is janky af and the total wrong way to do it
Are you doing the replace feature?
@icy kindle I have DMed you, written you an email. I wanted to see if you have received it. Thanks
what is ReMi?
If you have a question in regards to Suno, it is best to email support@suno.com
Messaging a random Team member, especially on Discord, would likely not garnish a response (as I would not be surprised if they get tons of DMs or e-mails for frivolous questions or items beyond their job description and ends up ignoring the DMs).
It is one of the two lyrics generation models, tends to be quite unhinged. The other model (Classical) being ChatGPT.
To change the lyrics model, click Write with Suno first. For a video, see #πβsuno-school message (note that the layout has changed from what is shown in the video)
I have attached screenshots for the process
does anyone here know a way for me to make a soft voice say something at the beginning of a song?
like it says the word / phrase before it starts
[Spoken Word]
I usually do something like [Intro: ASMR] (A test by Paimon)But it does not work all the time..
Maybe [introduction]? Though using [short catchy hook] before a message worked for me.
I haven't tried with [Spoken Word] at the start but that should work
In addition to what @silver falcon suggested, you can add descriptors to the tag, such as [Spoken Word: Whispered, Soft] - I've used that countless times.
Okay thanks!
does it only say it for the specific phrase or word
or do i have to type "normal voice"
after it
it will speak untill next metatag
are you swedish
yeah
ok
vad ska man lΓ€gga efter meningen man har lagt?
vad Γ€r det bΓ€sta alternativet
fΓΆr att jag vill inte att personen ska viska hela tiden
bara som en prod. tag
vill hΓΆra hur det lΓ₯ter
okej, men jag ska bara ha sΓ₯ att personen sΓ€ger "It is what it is" eller liknande
att det viskas som en prod. tag
sen instrumentalt?
aa precis, jag gΓΆr Hardstyle
[Instrumental] eller vad du nu vill ha fΓΆr instrument efter det.
HΓ€r finns en rΓ€tt stor lista med massa kul man kan anvΓ€nda: https://jackrighteous.com/blogs/guides-using-suno-ai-music-creation/understanding-suno-ai-a-comprehensive-guide-to-metatags
kan du sΓ€ga vad jag kan uppgradera med denna text om du vill kolla igenom snabbt
vad jag ska lΓ€gga sΓ₯ att AI:n fΓΆrstΓ₯r
den har gjort sjukt bra ifrΓ₯n sig redan, men sΓ₯ att man kan stoppa verses
o-o-o-o-o
o-o-o-o
[Set Reverb FX to ON]
Nothing is ever,
Take the chance
Enjoy the moment
Do not ever feel like it's over..
[Subtle laugh]
Enter the world,
The world of silence (silence)
The world that once was loud
Filled with sound
That I once remembered..
(remembered)
[Set Reverb FX to OFF]
[300bpm]
[drop]
o-o-o-o-o
The world as I imagined...
Is gone... (imagined)
The only way out is to imagine
(imagine)
[Slowly Fade-out Sounds]
vad ska man tΓ€nka pΓ₯
den har gjort sjukt bra ifrΓ₯n sig utan nΓ₯gra metatags innan, men vill fΓΆrbΓ€ttra om det gΓ₯r
Note that BPM tags don't work in Suno, as much as we would love it to.
Using more generic or musical terms (like Fast, Crescendo, [genre]core) are more likely to work
Ok
Well, it has upgraded the pace by a HEAVY amount when typed
i started by 180, and made it to 350 and it was super fast
jag gjorde denna med lite vocals och massa trams. du kan se i texten hur jag byggde upp den. https://suno.com/song/7f324291-38ba-4c24-bcdb-456cfc83abeb
hard house, trance, hardstyle song. Listen and make your own with Suno.
ok
synd att man inte kan styra hur AI:n ska lΓ₯ta
jag vill att de ska va pipigt i rΓΆsten fΓΆr att det passar min stil
kommer sΓ€kerligen framΓΆver.
sanningen
jag saknar mest att kunna anvΓ€nda samma rΓΆst till allt.
jag vill ha Nightcore blandat med en snabb rΓΆst
kan visa dig vad jag vill Γ₯stadkomma
lite sΓ₯hΓ€r
Bausa & Apache 207 - Madonna (Maxtreme Remix) [NXC, not my track]
youtube: https://youtu.be/__JTGmiZEc8
download: https://drive.google.com/file/d/1apTniF3uflrjD_RA0j1K6GJpbefqx8zc/view?usp=sharing
th
men som sagt
finns inget som kan efterlikna i ai vΓ€g tror jag
fΓ₯r pticha en rΓΆst i valfri daw och anvΓ€nda den som grund i suno
vad menas
gΓΆr en rΓΆst du vill ha och ladda upp den i suno
precis
okej
eller gΓΆra en cover pΓ₯ snutten
Could you use the metatags in the way you would use it
I want to see how I could use it
"Spoken Word" is that what the word is, or do i put the words after the metags?
[Spoken Word: Whispered, Soft]
(it is what it is)
[Spoken Word: Normal]
[Spoken Word] Det du vill ha sagt
aha okej
Jag brukar gΓΆra
[Intro]
[Spoken Word] Blabla
[Instrumental] eller [Verse] eller vad man nu vill ha efter pratet.
there are numerous ways to do it. You won't normally need a tag to "reset" the voice - without any additional reinforcement of the spoken word tag, Suno will be inclined to switch back to singing at whatever it feels the next logical break/new song section is.
kjan du skriva det jag skrev fast med metatagsen
ok, can you put the text as you wouldve put it
for an intro
I got no clue on how to put it as is, so I would like you to help on that part
sΓ₯ man kan inte lΓ€gga
"[blablabla: Whispered, Soft]"
[Whisper] blabala
fΓΆrstΓ₯r inte det dΓ€r
testa detta och se vad som hΓ€nder.
[Intro]
[Spoken Word] Yo! Yo! BerΓ€tta, Soulspin!
[Scratch] Soulspin
[Whisper] Jag kan viska
[Verse 1]
Vankar i mΓΆrkret, skuggan min vΓ€n,
gΓ₯r i tystnad, natten Γ€r spΓ€nd.
Gatorna viskar, hemlig refrΓ€ng,
It really depends on what I'm doing . If I want just a single spoken line, I'll probably try something like this first:
[Spoken Word: Assertive, Gruff] I like Cheese!
But this can also work:
I like Cheese!```
Or you can have "Spoken Word" and other related tags as the main styles and then play with some other tags to have more influence... Once you get the spoken intro how you like it, use that as Part 1 and extend from that point to bring in some music with extensions.
Lotsa different ways to get there. https://suno.com/song/547b778f-4662-435a-9eb1-1f8aed34888c
Spoken Word, Audiobook, Discord Voice Chat song. Listen and make your own with Suno.
... and I thought I was having problems with my ears π« I'm seriously starting to miss the shimmer to be honest. At least every couple of gens contained one with minimal shimmer and was nice and full sounding, so I could use that one. But now every gen starts amazing but rapidly deteriorates towards half way of the track. Gets insanely bright with the highs. All the gens are ruined because of this. I think I'll hold the gens for a while until this new issue gets fixed as well. Such a shame... π
Patience is your friend here π But it will get better im sure π
question... I wanna make some christmasy-sounding music for the holidays, but like... what would that fall under? "holiday sounding?"
like, is christmas music even its own genre?
@silver falcon https://suno.com/song/c11a0036-d27a-43c0-9e44-d1ca765d8531
euphoric hardstyle, hardstyle, fast, hardcore [female voice] [british female] [female] song. Listen and make your own with Suno.
Christmas is good enough for the genre. I even had Christmas Carol work
huh. okay. thanks.
Maybe add jingle bells or sleigh bells for that classical sound
I know there is no support here what I want is that the Suno team that we have in this server try to help.
I am paying to make songs not to waste my time and money if I was a free account I would not talk about it but I am paying 30$ for what no support and no help with the team that work with Suno
There are a lot of people using Suno and Support is swamped with emails. My recent email, sent in late Nov, had gotten a response in about 8 business days
If you Upload an Audio with vocals, the Public option gets disabled (a safety measure for people uploading copyrighted songs, which is against Suno's TOS)
You can still distribute the link to your song though.
ok
unfortunate.
i think this would be a good public song
euphoric hardstyle, hardstyle, fast, hardcore [female voice] [british female] [female] song. Listen and make your own with Suno.
It is something they did note in the upload
gg
Ok I get it that there is a lot people but for me itβs been 4days and still nothing and I know someone that needed to wait 2 months for a reply so I donβt know if they will reply to me or not but letβs see
But are they aware to that problem and are they going to fix it and is it happened to everyone or only some people the problems with replace section
Hmm just remember to account for business hours. They do not work weekends AFAIK
Two months without a response is way too long. If the person is on Discord, let them to know to DM me about that situation and I can forward that message along (which will get vetted first β many people said they waited weeks but it's been two days since they actually sent an email, in which case the message is not prioritized as it would have been otherwise in a legitimate situation like that)
I see and the guy is here in the server and Mike B helped him in the end.
What itβs mean AFAIK?
I believe they are aware of issues with Replace (the devs do monitor the suggestions and bugs channels) but I do not know if anyone is actively working on that right now.
Currently, the people in charge of the model and algorithm are more concerned with getting rid of major artifacts like shimmering first.
After that, it is likely v4 Extend would be next to be fixed (major issues with consistency) then Replace - though that is my guess
Ah, that was who I was going to contact, heh, as Mike B is the current Community Manager.
AFAIK = As Far As I Know
Oh nice
Ty finally some answers ty bro
This appears to be based on an upload that is either a pitch-shifted sample of this (at 2:08), or part of one of its remixes. Just a reminder that you must own the rights to anything you upload to Suno. https://youtu.be/HuQ1bldr7v4?feature=shared
You're welcome.
Mind you, that is just a guess on my part, aside from knowing they are still working on resolving the artifact issue (which one of the devs did mention spending time on the weekends to work on that issue)
Ok ok
And do you have any idea on how can I continue to make songs in 3.5 cos the v4 bugs on 3.5 too with the replace section so do you have any advice for me?
I assume you mean from an Audio Upload? You can crop the entire song (00:00βEnd) and the "cropped" song can be Extended into 3.5
Otherwise, if I recall correctly (IIRC), you can make the last Extended part in 3.5 and, when Get Whole Song, should have the entirety in 3.5 (but I haven't played around with that personally)
No not an upload itβs a song that I make in 3.5 but I need to fix 2words with the replace but itβs donβt work good already lost like 1.5k credits
Ah, that situation is such a pain, ugh.
For minor corrections, you should be able to use Remaster to fix it as well. It follows the lyrics closely and I was able to correct a minor typo in one of my 3.5 gens and the Remaster fixed that.
But I only did it once so I do not know how it is for replication, and well, Remaster makes the vocals clear but the music a bit more muted (or it did before, I didn't do it recently)
Otherwise I am not too sure personally - I haven't used Replace (or even Cover much). Replace and Crops are a pain for me because I am a mobile user. Sorry I can't help much with that π
Ok ty bro I will try that as soon as I get home
Itβs ok bro ty for the help
And one more question can I with the remaster change words and can add words?
If they are minor changes, it should be possible... otherwise it may hallucinate.
That is one thing people found out with Remaster that it follows the Displayed Lyrics in the song being Remastered.
People that deleted the lyrics had hallucinations. IIRC, one person mentioned they had to delete the song structure tag too, but I assume that is because they tried something that the AI did not understand
Ok bro ty
hello there, may you help me, how generate 10 songs at the same time ?
Click "create" 5 times in a row
Free users are limited to 2 generations at the same time (4 songs), paid users can have 10 generations at the same time (20 songs processing before it asks you to wait)
ho so easy ! thank you
and what about instrumental song during just few seconds, credits refund but may i do something wrong ? (in v3,5)
Not impossible that something you're using in the style prompts is causing that, but I'd expect that behaviour more readily if you were extending from the end of a song / silence
Is the voice from a persona fixed to the style of the track it was created from?
start fresh !
TY!
If you want to share the style prompt, I could see if I run into the same issue
no the song is 1:50 and it skips half the words
Just to update on jingles and the like. (random 3.5 or 4.0)
I simply started, the regular basic way with a new song (on 4.0) and wrote a couple of verses and chorus (1/3 of a song, no tags), and all songs out of 12 gens (ummm 6 gens... 12 outputs) sang everything once and cut off between 1:04- 1:30 without any repeat verses, instrumentals or hallucinations. The pieces didn't close like the jingles did, so this is different, but just saying, all the other past and usual "continuations until it reached 3 to 4 minutes worth" was not there either.
Planning...A new series of tests: 1) testing short songs, from 15 second jingles to regular 1 - 2 minutes, a couple of ways, for getting a solid button or fade ending.
2) testing if most songs now will end or at least time out just after all text has been exhausted once.
2) making sure hallucinated babble-scat, gibberish songs are still possible. I'll be sad if they are no longer.
Any related feedback is welcomed to add to my quest.
Are you using the "Exclude Styles" option for chimes cymbals synth wave, or did you type the minus symbol into the "style of music" section?
