#🗣️┃chill-chat
1 messages · Page 21 of 1
Suno can give some really good samples sometimes. I've prompted it for things like drum loops before. Good way to integrate AI into your workflow. But, as you're just starting out with it, have fun experimenting 😄
what do you think of my newest udio track?
I'm always amazed at how well AI can create music.
I'm even more amazed at things like this where we're able to bend a music generating AI into doing other things
thats my main goal... sound like something that doesnt exist instead of trying to be like someone else... and thats why ai music is so cool for me. only my imagination is the limit basically. i dont have to buy fancy snths or vst plugins, learn them and exploit them.
In case you missed it, I made this a while ago. I ended up calling it "The Moisture Demon" and have been generating a LOT of really weird, unnerving video clips with Noisee to go along with this: https://discordapp.com/channels/1069381916492562582/1159929934270181437/1228156146142023700
Kinda my favorite part
https://www.youtube.com/watch?v=TZjjCpXXKqU alan walker using suno
Listen to my new song 'Unsure': https://alanwalker.lnk.to/Unsure
Do you want to be included in the official Remix EP? Read below.
If you're a music producer like me, you've probably wondered if AI will ever replace us in the future.... I decided to put AI to the test. Can I use AI to recreate my own song? Or better yet... can AI finally allow m...
i found the zombies
can someone suggest an ai tool that is good in generating music videos to my created songs?
This is incredible, great creation!
First minute sounds like a 2000s pre-opening to a movie.
Who is the better mascot for my lofi channel:
Worried Astronaut or Cat in a Submarine?
I think the cat.
It's the internet. People like cats.
Also the picture is brighter, and the space-age-looking submarine seems cooler.
🟡Ready to become your own record label? Get started now!:
https://www.topmusicattorney.com/becomeyourownrecordlabel
On this segment of the Top Music Attorney Podcast, Entertainment Attorney Miss Krystle and show/music producer That Orko delve into the nuanced world of copyright law and the growing influence of AI in music production. Inspired by...
cat. But fix the eyes on the cat. 🙂
the cat
I think people have success with those
I am also with the cat.
Because firstly Cat.
But more importantly the worried look on the Astronaut one doesn't really fit in my opinion.
But something about the cat feels a bit off as well. The way it stares right at you with this weird look... But maybe thats just me 😅
You'd stare like that too if someone put you in a watertight pet carrier to visit the underwater vet!
I was paying attention to that one at first in hopes of getting some basic information, but they have no idea what they're even talking about.
"Scrubbing the internet" "Scrapping" not scraping. There is a difference.
And this is supposed to be a lawyer? Their whole bread and butter is supposed to be words and meanings.
They can't get the basic terminology correct and we're supposed to take them seriously?
Oh and be sure to join my course to be your own record label...
I see classic YT "shock pose", I don't click
Probably good advice.
Also the text looking like a tabloid headline
I tried to give them a chance though because no one else is really approaching the subject from the legalitities aspect, but the more they talked the more obvious it became they don't actually know anything about the topic to which they are claiming to be speaking on. And the producer dude is just as bad.
Instead of trying to understand or consider legal ramifications all they do is spread misinformation and act as a lightning rod for others to also spread misinformation and be mad.
It may do wonders for their channel but I really really wouldn't want anyone like this to be the face of any legal concerns I had regarding AI and music.
would NOT want this lawyer defending me in court.
Just sayin
But..but...she has purple hair!
And librarian glasses!
Surely the best dressed lawyer is the winner, right?
😅
Does she walk around with an OOPSIE face?
¯_(ツ)_/¯
I wonder what instruments to use in an intro
I don't think the judge would care for that
(I am an expert. I've watched Perry Mason)
what is the genre?
Classic Mason or the newer one?
ugh! you have to ASK???
dark new wave
LOL
🤔 I have been trying
[my genre buildup]
or
[my genre intro]
so yours might be
[New Wave buildup]
or some variation that references the genre, and Suno should make something appropriate
Ah I did not click the video because I was expecting stuff like that...
Is it that nonsense again? Like with Image AI ripping apart artworks from the internet in realtime and pasting it together to generate the output?
Not so much the real time bit but otherwise yes.
yikes...
Along with a huge heaping serving of FUD.
Unrelated to cat, but your suggestions on managing the static/hiss/metalic tones on generations by using the tags you suggested seem to be working excellently for me, many kudos to you
Cat on an Atomic Surfboard....?
Neon surfboard
I am so exhausted by these type of people.. I have started to just quit any online discussion/video/other content that has a negative view on AI which is probably not good as well because I miss the true counter points but it is just sooo much nonsense I can't..
Did anybody say cat?? https://suno.com/song/33307ef0-62de-41ee-a748-236d219796e1
Female, Hip Hop, Pop, comedy, Scratching, stutter, Monotone vocoder robotic voice, urban sounds, Memphis song. Listen and make your own with Suno.
Honestly I'm just tired in general of everything being clickbaity or desperate engagement attempts. Modern internet is really bad in that respect.
It is the tradgey of our times that we no longer can argue with reason and are surrounded by emotional appeals.
Neither side is great at it tbh. And both get obscured by drivel and the actual good points never get attention.
a moderate, the worst kind 
I think I only had 2 discussions so far that where an actual exchange about this type of stuff where after the discussion both had a changed view on the subject matter.
Anti AI is this ~moments~ 'remix isn't real, digital art isn't real' thing
But it is just that, a moment
It's a tool that is finding its way into people's hands
Especially when it always gets to the real argument: How do people get paid
Valid, but someone is, even if they are using AI
The idea that executives somewhere are typing prompts into LLMs is absurd
I'm not done complaining that Johnny Cash never went to prison.
(total fraud lyrics)

The real problem is
But nobody wants to talk about that so instead robot steal my job no I will not adapt
Gotta laugh when there's an app with both Suno and Udio in it's title and the images for it are just nonsense like this (not installing it, just browsing around)
@umbral wadi i tried previous to link a utube video, of the fake udio music stuff, all i got was aload of abuse here, so i don't bother with that anymore, it just doesn't seem to be worth it.
personally thou, i do thing that a voice training person would know the difference between fake copies of artist and real people artist voices . (just saying)
@late crown
yeap, don't know why but udio keep on changing the prompts to movie track:)
Ah I always turn that function off I want to see which tags do what 😅
got to remember to do that next timee:)
Hi everyone i want to use bot suno for generate my music but i dont know how
it resets every time you reload the page as well quite annoying
Chirp bot is retired, use the website https://suno.com
Some helpful starting links: https://suno.wiki
https://youtu.be/lj8utBd_IUM
ok but i having isssu and since monday last week i contacting support is never have reply me
I know, but I was wondering what was going on
it old that why? has suno has updated to v3
Yup Enemy of all, friend of none.
hell yeah
Now we just need the ability to upload custom images to Suno but I also suspect that won't be exactly how it works
Is there any way to train an IA and then they cover our songs?
The songs we made on here I mean
That's always the rough general idea of the guy
Past his prime, just chilling making music on his own
Insanely talented though
Well not on his own, but for himself
how do i fight a vampire
Honestly it's not that bad 😆 ... she used the wrong word when she said "scrubbing", and the guy used the wrong word when he said "scrapping", I don't see that as a big deal. I mean they should know the right words, but scraping isn't generally associated with traditional music so I can see how they might be less familar with the term.
I watched the original Udio video she did and this one also. Despite the clickbait cover I think the information is pretty reasonable. There's nothing that she says that most people here don't already think anyway. The clickbait covers are definitely stupid though 😛
nah nah nah nah it doesn't even get the basics right it's complete bullshit
hmm i don't really agree, otherwise there's lot of people you wouldn't listen to because they are uninformed on some specific part of something
Exactly yes
if this "specific part" is some very basic thing about the stuff it only shows the person does not really care about the technicalities and follows a different goal
Yeah, i hear ya. I just don't think the video is that bad after watching the whole thing
any good apps to create an AI voice?
Is it though? Or is it just a "well made" opinion piece without any real basis in reality? Do they understand how the AI works? Do they know what they talk about? Or is just an ad to become your own record label?
it's just a pretty good "basic" video on the topic (the original Udio one she did was a more thorough one). I'm honestly happy to watch videos coming from people in the music industry that don't say "AI = Bad". We need as much of this as possible. It's just a pity from the video cover it appears negative lol
in programming, i remember a lot of people were anti-AI and then there was this slow drift into acceptance, and now nearly every programmer uses AI autocomplete. You can kinda feel this happening with music right now, and these neutral/positive videos help that
Hm..
still off-putting. not my kinda content with the clickbait and the lack of research on terminology. It's an important topic and if such errors occure and are not even noticed by post production and corrections edited in I can't trust the rest to be researched as well and I have to check everything myself which defeats the purpose of watching a video..
But I am also really fed up with stuff like this so my opinion might be stronger than necessary
understandable! I have a feeling you probably know the "lay of the land" with this stuff anyway (well I'm just guessing), the video isn't really saying anything new
yea those clickbait youtube covers are definitely a bad thing lol
instant turn off. On Desktop browser I have an extension that changes the thumbnail to random part in video 🤣
That sounds incredibly useful lol
I DON'T KNOW WHERE TO GOOOO
what's that Kits AI thing, isn't that for voices?
but we can't create a voice, can we
What you mean "create" from scratch or like train it on your own voice for example
not sure haha, i just know Kits AI can do something with voices 🤷♂️
I have tried few times but couldn't make music there
What is your goal? What do you want the voice to do?
I want to create a voice '^^ for my character haha
actually I want to make the same voice singing the songs I made so far¨
So this is possible but rather complicated if you want to create a voice and not clone one. And it will probably not have the best quality.
But as far as I know on eleven labs you can edit a voice in a kind of voice creator but I don't know how much can be tuned. Then you could use that voice to generate a lot of spoken text and train an RVC model on it and split your songs with UVR and convert the voice with RVC. But as I said this is not easy and quality will not be great
I would start by seeing if eleven labs Voice Editor can achieve a voice you want
I have made the songs in Suno and own the rights but idk how we can change it
You will have to split it into instrumental and vocal stems. Then you can convert the voice with yet another AI. But you first need the Voice. And it needs to be one you are allowed to use so either permission or an AI voice
how do we split instrumental and vocal
UVR
give me a second

or maybe should I clone the voice I like among the ones I have made so far
Do you mean in Songs from suno?
That will be very very hard to get something useful out of that
yeah that might be too hard
look into eleven labs voicelab
See if you can do a voice there that sounds like you want. If that works I can see and help with the rest and maybe it will work but I wouldn't hold my breath for something amazing this is a really complex experiment
If you have a longer song with a voice you like you are comfortable with sharing you can post it or dm me and I could see if it produces something useful
I found a lovely voice in there 😄
You can try if the elevenlabs voice changer already works good enough. Should be somewhere on the site as well. Just make sure the voice can be used commercially.
Hmm now i kinda want to see if I can make a voiceclone out of a suno song... gonna try tomorrow with some of mine 🤣
If it works good enough I'll let you know
how to change the voice
on elevenlabs? Or in general?
yes on eleven
Search for the Voice Changer.
Should be there somewhere but it might be that it can not really sing
/chirp what happenedddd
bot is no more
you have to use the site (suno.com)
excellent! Glad it's working, it seems to have helped several people thus far, so presumably it's not only applicable to my particular genre
which is great, I'm sure there will be edge cases, but likely the same concept, namely choosing noise heavy tagging, will help.
I am trying DUBBING
@graceful swan not really singing that's so weird haha and creepy
singer-songwriter xyz soloist
has served me well also.. just a gamble if you get your actual prompt with clean vocals or a happy sing along with acoustic guitar but that's expected I suppose
yea, that one also works pretty often
the ones I've been using lately are insanely clean, but they really slow it down
so you have to have a good intro
like, I have a lot of gens I didn't put put bc too boring
but the voice
The reasoning is pretty sound too, awesome you figured it out
Ah well than there is probably only the RVC way... I will check tomorrow if I can make it work with a suno song as only basis but I think it won't work.. quality wise.. don't have many other ideas. You could make the eleven labs tts speak a lot and train RVC on its output but you would have to read TOS it's probably not allowed
post-experimental(neo-bachzarthoven indie post-anti-gothline){Avant-garde post-post-vocalization}, cassette fuzz lofi fx song. Listen and make your own with Suno.
this was it
I have so many clear gens
I need more space in the style prompts 🤣
but I ended up deciding I needed to go back and totally redo
but with an energetic instrumental open
it's just too... meh. I know it can be better.
recently i have slapped "polyphonic violin virtuoso" at the end of everything. Quite interesting how strong that is. Eats a lot of prompt space though and without polyphonic its not as great.. epic works in a pinch as replacement
Signs · Spock
Signs
Released on: 2022-09-09
Copyright © Christopher Mikhaili All Rights Reserved
@graceful swan they give me only posh accents
give it a shot if you want @jaunty thicket
its suno made ?
nope before
oooooh
made a year ago
gimme a sec
ultra
will speed it up further
ima have to slap one real quick
before I go back to work
One sample
https://suno.com/song/766a4047-267c-49f9-be0f-b6fbbcf06e7d
Almost instrumental sample
https://suno.com/song/a9bdd719-c772-4d81-bdcd-c17f1dde53eb
fast paced, melodic female singer-songwriter soloist violin reggae, death metal virtuoso, polyphonic violin virtuoso song. Listen and make your own with Suno.
👀
nice
its very different than what ive seen some people post so no problem if nothing can be remixed
xD
bro
Hm maybe it's a dead end then.. Will check my other idea tomorrow and report back
thats a good answer
omg that beat
rapping over a video-game level
my character is british but not posh
I EVEN ADDED JERSEY
gimme da brogl
LISTEN THE JERSEY PART
got the virtuoso part at least
I really like playing around with that. It's really cool how it can replace the lead guitar in metal and stuff with a violin
tends to confuse higher pitched vocals with a violin though and vice versa
Ahahaha... Mouse makes a difference 😄
yea, I've had that in interesting ways
where you can have a vocal transform into a guitar solo
hmmm, why i cant get the video like before?
reminds me of the cat i use to know, that was half wild and only came in the house for food, try to stroke it and you would of got a nasty scratch, trap it in the house to and the same it will scratch you to pieces, the cat was an abandon has a kitten, and we took it from the shelter to give it a home. but would it stay in the house, NO!!!!😆 we name him "FROGGY"
If it does it in the right way it sounds awesome
@solid vector we got something here
niceee
cant wait to hear it
I will share it on my YouTube if you want and credit you of course brotha
working on the hit hat rn
can we talk about the breadth between these two hahaha
ultra polytrapic-mathtrap gothello-gothline, chaos cassette noise-glitch virtuoso song. Listen and make your own with Suno.
ultra polytrapic-mathtrap gothello-gothline, chaos cassette noise-glitch virtuoso song. Listen and make your own with Suno.
Sometimes I get really good cover images.
Hope it does not return back to birds 😄
i have a question. the songs we make using suno. are we able to use them on things like youtube or twitch? or does suno own the copyrites. i havent the first idea on who owns what legally in this situation. i assume its suno. but any info would be great.
If you are a paying customer it's yours to use however you want.
Answered on the sites FAQ/help section
didnt evan know there was an faq on there sight. oops. also thanks for the info. ill try fing it. also paying 10 a month. so yey mine.
Hello! Please email billing@suno.com if you are unable to find the answer to your question on this page.
you are all awsome thnk you so much.
one last question. is there any kind of way of seing what was generated as free and what was generated as payed. as i did some prior to throwing money at it lol. if not not to wurry as i dont do streaming or youtube often at all. so could easally whipe nad restart now im paind.
No other way but to look at the timestamps currently
brilliant that helps. didnt think of that. 😄 thank you so much for all the hlep and advice peeps. 😄
my fav rapper
rhodes AHHHH
@solid vector recording
tell me when you are here
im here 😄
i cooked
I love that you chose the guitar outro
bahahahah
@graceful swan do/did you use reactor with davinci?
but ima maybe drop it on my main channel with your name
witheout the rapper

no not yet
besides noisee, what is everybody's go-to ai video generator
looking for something as simple as noisee, but i'm a boomer so whats the best
Not sure what is the best since I just used Noisee but I linked to two YT videos that talk about other AI Videos at #🗣️┃chill-chat message
Edit: the video links
https://youtu.be/mKcfo4zbcUA
https://youtu.be/uoWOzWzWC7U
📄 Download free AI eBook here: https://clickhubspot.com/y6k
Links to Tools:
Viggle - https://www.viggle.ai/
RunwayML - https://runwayml.com/
CapCut - https://www.capcut.com/
Noisee - https://noisee.ai/
PixVerse - https://app.pixverse.ai/
Haiper - https://app.haiper.ai/
Pika - https://pika.art/
More from Futurepedia:
⚒️ Get recommendations on t...
Check out Careerist today:
https://crst.co/7T1vW
Save 600 with code: SAMSON600
Join me as I dive into the world of AI video generators. I'll be testing and comparing the big players based on speed, quality, and price.
📧Join my newsletter
https://delightfuldesign.eo.page/w7tf5
👨🏫Check out my AI courses:
https://www.udemy.com/user/samso...
Found this YouTube short ad about an AI music Gen that shows SUNO's music video thing... Is this legal?
unless if I'm seeing things, but that's clearly the video that you get when you download any of the songs from SUNO AI
There's a few sites and apps that basically scalp the service and do god-knows-what with your information and payment details.
Ah, okay
They really shouldn't be doing it as I'm pretty sure it violates the Suno ToS (I struggle to find the specific part but I remember it being talked about ages ago)
g.night all
old school hip hop song. Listen and make your own with Suno.
...awww where'd the kitty go?
steal dah kitty
Here's a cat for you
I'd like my kitty wireless, please
how sweet would it be to have a player that you could load Suno tracks to - that would play on this that would "animate" to the beat
i mean a physical thing
like the futurist version of our Cat & Fiddle animated clock
anamatronic clock i should say
A look at Bandai's Little Jammer Meets Kenwood 'Live Stage' music entertainment system.
MERCHANDISE
Techmoan Merchandise is available from
https://nerdkeyz.etsy.com/
SUBSCRIBE
http://www.youtube.com/user/Techmoan?sub_confirmation=1
SUPPORT
This channel can be supported through Patreon
https://www.patreon.com/techmoan
*******Patrons usua...
gong sound JAPAN
https://www.youtube.com/watch?v=s8It2O1RdG8 but small & in your house
Filmed at the Cat and Fiddle arcade in Hobart, Tasmania. On the hour, the clock plays the children's nursery rhyme.
the fact that i was obsessed with music boxes with moving parts as a kid has not changed, i even built one for myself that is part music box & a small model merry-go-round
So I just learned about kits.ai from a YT video I came across (for splitting vocal tracks and swapping voices). In the example where he combined (in Audacity) the initial split voices audio track combined with the gender-swapped vocal of the same song, he had a very great duet made. So many credits could have been saved in my duet attempt
https://youtu.be/5jseFmgAP2M?t=9m14s
- 6m 10s for importing of a Suno song for vocal splitting
- 9m 25s for the combined duet track being played in Audacity (original track had the female voice, he converted it to a male track in kits.ai, and placed both in Audacity)
Sing with the voice of any singer, and even Master your creation with Kits.ai
You can start totally free, like I do in this video!
There are many practical reasons for a musician to learn about voice cloning and conversion, and Kits allows you to experiment with the vocal and instrumentation of your music in a ...
Oh, it has been a month since I joined this server. Time flies by fast
we'll try
IN offtopic because not suno - but i feel the inspiration of darkness, thanks ireland: https://www.youtube.com/watch?v=BNc5zTYkTaQ
Subscribe and 🔔 to Eurovision 👉 https://www.youtube.com/user/eurovision?sub_confirmation=1
Bambie Thug from Ireland performed ‘Doomsday Blue’ in the First Semi-Final of Eurovision 2024 in Malmö, Sweden
-
We are #UnitedByMusic. The Eurovision Song Contest celebrates diversity through music. Please keep your comments respectful. We will not t...
https://www.youtube.com/shorts/X3MeXe3fDn8
This has Meme Song potential 
@frozen temple this started on my soundcloud and it remebered me you lmao
https://soundcloud.com/geno777/clown-freestyle-ft-prince-rose
Dm on ig for feats etc.
I like his not-caring style.
bahahah
ima use this on my next remix of you bahahah
lol
https://arxiv.org/abs/2404.09313 open source is slow to catch up
A song is a combination of singing voice and accompaniment. However, existing works focus on singing voice synthesis and music generation independently. Little attention was paid to explore song synthesis. In this work, we propose a novel task called text-to-song synthesis which incorporating both vocals and accompaniments generation. We develop...
Anyone tried inpainting in Udio? Looks lime they put it behind the subscriptions.
Suno needs this fast, falling behind.
I honestly haven't touched Udio for a while because it doesn't produce the style of music I liked despite trying several times 🤷♀️
My wife has been handling my Instagram so I've been able to focus solely on SDXL gens (while I'm nursing my suno credits).
Thought some of yall might like:
Just creeps me out 🙈
i wonder how many people on this server are music industry plants with the sole intent of producing cringe content and responses from people
it's made using a LoRa is designed, it'd actually base SDXL
not me obviously
where I targeted areas of overtraining
you're a plant for some other entity XD
This model is based on work I did on my "Unsettling" Lora. It uses some of the images generated there, along with subsequent images made using, aga...
I am a flower
can someone help me?
help you escape?
ok ok, you're at least not a bot!
haha
nah but fr, what you need?
0
and one person's cringe is another person's grunge
we could be in the "then they ignore you" phase vs the "then they fight you" phase
i tought i have problem but i dont 😄
4
Yolkhead fixed it by observing your message
schrodinger problem
nah, they will fight with litigation, botting creates engagement
do like, it would benefit suno. There's a name for it I can't recall, where you create bad attacks on a brand that are clearly intellectually dishonest to create an opposite impact
Suno isn't doing that, to be clear
but anyone trying go bot runs that risk, since the service is genuinely good
it's what happened with gpt-4
the more bots repeat the same 5 phrases that anyone using the service is immediately confused about, the more brand loyalty they create
which, incidentally, is bad for consumers, but that's neither here nor there
I just realized for some reason Discord uploaded gimped baby thumbnail versions, but oh well, you get the idea 😆 if anyone wants art ever for cover stuff or just other projects, unless I get inundated, feel free to reach out as I like making it (ie I'll do it for free, obviously)
yeah birds are boring.. I believe they are working on more freedom on your images
but I'm talking about more profound weaknesses. From what I hear so far Udio is better in most regards quality-wise. I thought Udio doesn't allow to "continue from anywhere" and that was the only thing stopping me from checking it out (no creativity if it's not an option)
(obviously it wasn't an option in Suno too, first, but I never looked back)
how is Udio in terms of lyrics adherence, especially for continued generations?
Just got back to the computer
I see Udio have just announced their subscription pricing.
@maiden hornet to answer your question from #💡┃suggestions-web, i'd say the main weaknesses of Udio are shorter generations and slightly longer time to generate. It also just has different interpretations of certain genres/key words which could be a good or bad thing depending on your taste. Basically the same prompt between Udio and Suno can have vastly different results.
Are the credits the same?
cool, thank you. Did you try any heavy genres, I'm talking metal/core?
1 credit corresponds to 1 ~30 second audio track. Each generation currently generates 2 tracks, so each generation (including remixes/extensions) costs 2 credits.
oh, then that's much more than what Suno offers
Suno 1 song = 5 credits, so 10,000 credits means 2,000 songs. Udio 1 song = 1 credit, so 4,800 credits means 4,800 songs. Basically 2.5x, which is a huge difference. I wonder if Suno responds by lowering the prices.
My Suno subscription ended today so it feels like a good time to try it out
@crisp fern , side note, so you actually squatted suno.com/@songs lol? 🙂
Keep in mind.. they have 33 second per generation
yeah that's the least of my issues. I usually cut my generations in 5-10s at a time anyway.
it's very rarely that I need the entire 1 generated minute (only if the AI is having a really good day)
I also feel user control is a bit better with Udio in the sense that it is much better at listening to things like metatags and won't ignore lyrics as much. It's hard to describe but Udio has a bit more of a raw "garage band" type feeling where it doesn't try to create a concise composition as much as Suno will. Which can be good or bad depending on what you're trying to achieve
Progressive metal yes, sounds pretty good
Yes, but you comparing then as if it's the same, which is not..
sure sure. I just mean, in my mind, 1 generation is 1 generation, I don't care that much about the length. I will spend ~1000-1500 "Suno" credits on any given song
just happened to be online when they made the handles announcement so I quickly grabbed it lol
care to share an example?
and the best I could come up with was my name lol. Well, anyway.
With Suno you can continue from any point, there you are stuck to the 33 seconds, and your continuing are always 33 seconds too. In paint is not going to change that
Maybe instead o lowering prices, keep the same just increase the credits or make credits cost 2 per song instead of 5?
yeah but I think @crisp fern said you're not actually stuck to the 33 seconds — you can continue from any point just as well? If not, that's a deal breaker for me
Yeah, well, that's what I mean. Shall we call it unshrinkflation?
Well if you are going to use it.. keep us posted!
perfect xD
I will rubs hands
And as I have your attention — can I post a song here I'm very proud of?:)
lighter than most of what I do with GENERATED, but I love the Shinedown/Godsmack feel of it https://soundcloud.com/vzkrv/mira-is-going-to-the-sea
Based on my sister Sasha’s eponymous, unpublished short story. Spotify: https://open.spotify.com/track/7vkp28QDkOZ6bcMIUi6LkA
She wakes up to find scales under her pillow,
Whispers to the mirror, “Is
(Sorry it's not a Suno link — I made the ending part by extending the very first chorus and glued it together in Reaper, I often do this because Suno struggles to keep the riff unchanged after the bridge)
Can you imagine what this will sound like when the sound quality is better.. Holly sh*t
offtopic (as due) but did you watch the new Beavies & Butthead? They rock
I cannot. I mean, it will obviously blow my mind even more, but it's hard to be imagine how much 😄
This one is a little more tame.. but if the quality was better.. OMG..
https://suno.com/song/ab5faecd-a16b-4b27-8156-ff9c4f63f867
Art rock alternative rock electronica experimental rock, build up song. Listen and make your own with Suno.
love the bass sound
we need a minipoll on how long we think until the first lawsuit against Suno
I dunno, maybe some generation with a very definitive vocal texture, like, I don't know, Mik Jagger's
Or if the AI accidentally reproduces a whole sample from somewhere. Highly unlikely but not improbable
I heard more things that I found suspicious with Udio. Not trying to diss them, just a fact.
Audio inpainting with Udio now eh?
I'm guessing the models are the same more or less, math-wise
Bringing the "a" to udio
My phone INSISTING I mean to type audio every time
At least it's stopped trying to correct fuck to duck
Maybe.. but they have parameters. If you give more access to underlying, and it's copyrighted, it could become a problem I think
sad that its behind paywall so so not gonna test it
Just viewed the demo . You can change a word in the lyrics or add à piano solo in the middle of your song . Incredible.
lolol. It was only in like two weeks after hearing the name that I even realized it comes from "audio"
I mean they have to charge at some point
can you post the demo here?
Yeah, I don't trust free services that much anyway
by the way, what are Udio's terms in terms of commercial usage? More or less the same?
yeah but releasing a new feature and putting it behind the new subscription is just a move I cant get behind. I mean sure from business standpoint I understand will bring in more subscribers that way but ehh.. If I cant test every feature before subscribing I am not subscribing
Go in this song to the 35th second:
https://www.udio.com/songs/tha5q6AU2pf43bs911oXfe
@maiden hornet
I'm not very well versed in female rappers. Nicki Minaj I'm guessing?
(I was assuming you were referring to the vocal texture similarity discussion, just in case)
It's interesting, from a business looking POV, that their generations are 30 seconds a pop. Makes it difficult to cross shop Suno and Udio if one were to try
Also illustrates that the models probably have a fundamental difference between them
why?
Because people have this idea that they have to use "all" of their credit
With Suno you can get UP TO 2 minutes per gen
Udio is 33 seconds
I don't care because I'm after entire songs so whatever
It does look like Udio are following Suno's subscription model quite closely. Each gen gives 2 outputs, which are now 1 credit each instead of 1 for both total previously. Free users will be able to press the button 5 times per day with an additional 50 times per month (which is different to Suno)
But customers try to like, get the most MPGs
But you'll get half - or even a quarter - of the output
Ah, ok. As I said I usually create songs at 5-10 second increments at a time, so whether it's 30 or 80 doesn't make much difference to me. But I see how it could be different to others
So again, looking at it in the bag bang for the buck lens
Ideally in the future we will be able to have more of a dynamic option to choose how long the output is and it can use more or less credits accordingly (for either Suno or Udio)
There are people that want to hit a button and have a complete song pop out the other end
Yeah I'm like you, I just want what I want at any one time, which is why inpainting with Suno has been a top want for me for a long time
I still think music AI is super cheap, in general. I mean, 1000-1500 credits per son is like 3-5 bucks. It's basically tap water
*per song, lol. Sons are much more expensive
That's why I have no qualms over throwing credits at something to try to get a better generation.
In context yeah, but if you zoom in to JUST Suno and it's interface 'credit' takes on a different meaning
People are bad with money
Extra bad with pretend money with blinders
Yeah compared to AI image generation you definitely get more bang for your buck with audio it seems
at least comparing a $10 midjourney plan to a $10 suno plan
images I can generate locally though. Audio not yet
Midjourney has gotten really good though, like eveytime I need an image I can seem to pluck it out of the generative furnace that is their LLM
It looks like it works out roughly the same price for about the same amount of time/credit, just doing some quick calculations.
jukebox
Certainly by design
It's... well, it makes sense.
Yeah it's honestly insane how much it has evolved from V3 to V6. That's why I'm hyped to see where suno goes
Also this
show me a jukebox song that sounds close to suno quality
with vocals
this is the highest quality I was able to get with Jukebox https://www.youtube.com/watch?v=SfiIrcIoz5o
For the Generated project, I create music using OpenAI Jukebox. I prompt the model with a genre and artist reference, it generates several 3-second samples; I pick one of them, it generates 3-second continuations; and so on, until you have a finished song.
This track is part of a to-be 5-piece EP called “ONTES” (Ouroboros, Nihil, Thanatos, Ereb...
no, not close. but no local model is close to midjourney either is it? not too well-versed
a lot closer than suno and jukebox I would say
You can get perfect stuff out of Stable Diffusion locally
It just requires you to reach in
SD running inside Photoshop with masks and layers and all that is fantastic
Afaik jukebox is kind of different because you can actually feed it an audio file to extend/continue but I haven't messed with it in like a year
And has been since 1.5
you're probably right. I think it's because (a) most people are much more into pics than music, and (b) music has a much more powerful lawyership
you can with suno too i'm sure, it's just they don't open this to us. But this is how e.g. "continue v2 generations with v3" works (I imagine)
in my experience diffusion models are not up to par, creatively
If it's anything like Stable Diffusion then
Or hear me out. Because AI music is the bleeding edge currently. Will take time for Opensource to catch up.
Always like this was so with AI images, TTS, Voice Clone. Will be same with music
I come from that art side, professionally, so the models don't have to give me complete images directly
Using it as a generative fill is plenty to do whatever I need it to do
That said, Midjourney is clearly the winner
As close as we can get to one shot prompts
Couple that with local SD inside Photoshop and there really is no limit
@viral flame still one of my favorite intros
https://www.youtube.com/watch?v=CSvFpBOe8eY
Official HD music video for ”Chop Suey!” by System of a Down
Listen to System of a Down: https://SystemOfADown.lnk.to/listenYD
Subscribe to the official System of a Down YouTube channel: https://SystemOfADown.lnk.to/subscribeYD
Watch more System of a Down videos: https://SystemOfADown.lnk.to/listenYC/youtube
Follow System of a Down:
Facebook: h...
I quite like that Udio show the per-month sub price first. Less likely to accidentally sub for a year thinking you're paying per-month I think.
An amazing song, been one of my favorites for 20+ years.
I was referring to something more like this:
https://youtu.be/UrIiLvg58SY
No disrespect or anything. Great song, great band.
(But how is it classified as metal?)
REMASTERED IN HD!
Official Music Video for "More Than Words" performed by Extreme.
Listen to more Classic Easy Rock songs: https://www.youtube.com/watch?v=NlwIDxCjL-8&list=PLjF50Dlp9iem_B-VeFMNvIAVxgrixTIPu&pp=gAQBiAQB8AUB
Watch more videos Remastered in HD! https://www.youtube.com/watch?v=ub2gw0nIRTo&list=PLDNtAuXIhbEPLcw6HLBLkVJl_MUd0DFW2&pp...
yeah that's very unusual for any SaaS, but I appreciate that
I sing that song when the stereotypical long blonde dude in the group pulls out their guitar lol
but you're not wrong
heads up, not only you can select a section, you can even choose it on a waveform, with (I imagine) sub-second precision
alas, no commercial usage, it seems. Maybe they'll change that now that they introduced subs
Basically you don't have to pay to use your songs commercially, but you must credit Udio (section 6 of the ToS IIRC)
oh, you're right, I'm wrong @graceful swan
right. I mean, I don't mind, I never hide the fact that I use AI. (Basically my "main" project name is GENERATED lol)
I'm okay with that, particularly as I'm a relatively light user of the service.
Is it possible to remove the little rocket?
Nvm it's attached to the role 
Can't get screens to be fully in view in the background 
But I do like her.
Very funny.
guys i need 3 more likes on this track to get it on trending so please like this if you want https://www.udio.com/songs/93FBbMwnc7N7wjWgE9rbv6
im going to make a song about drinking petrol from motorbike has i discovered a leak in the fuel pipe 🙂
oof
who likes eminem
One drawback about Udio (coming back to the earlier conversation) is that it doesn't stream the generation, so you need to wait for it to be completed in full before listening. (Like back in Suno v2 days)
@dusty patrol there's Off-Topic
and then there's spam.
Have some time out. Maybe you can find the Udio Discord.
funny fact but my most listened artist this year is my own project haha
I kinda wish their extensions would play back at the start of the new section too, kinda annoying to have to click around
a good thing for me, I'm tired of gluing stuff in Suno
hm, do you know how I join it all together in Udio? Seems like it glues the current extension with the previous (or next) part, but not all in one song
It's nice that it glues it, but I'd prefer to start playback at the point where it's glued, or close to it
oh, yeah, right, I see
i perfer the suno way, that way you don't need to redo a whole song. if it fails in the middle, then you can just re do the part clips.
without the noise Suno would be so much better...
it would be nice if we could have a tool to remove it
topic suggestions for my first song
goats? 🙂
Salut, j'ai fais un son de techno/hardtechno melodique avec ma guitare pour la mélo si ça intéresse 😉 https://soundcloud.com/kyusenne/magnolia
Hardcore,Techno, Hardtekk, Uptempo, Raw, Frenchcore, Industrial, Minimal, Hyperpop, Frapcore, ect
https://www.youtube.com/channel/UCYcp2Uhe7f5dm_DaxvIf4fQ
wow, Udio actually remixes when you say "remix". So say if you want to add an instrument, you add it to the prompt, click remix, and you get the same tune with the new instrument. It's mind-blowing.
@late crown tell me if I need to stop this
Found a 40k space clown in my backlog
also, a feature I long for on Suno, tree view.
hopefully this will also serve for the team to look at the differences and maybe add some to the backlog.
things I hate about Udio so far:
- no streaming generations (you have to wait)
- it always jumps to the next generation after playing this one, no option to turn it off as far as I see
- can't generate several extensions at once, it throws you back to the list view once you click Extend
The features Udio has are nice, but for me at least it seems to forget half the prompt.
a wild @fast rivet appeared in my klownpunk prompt
Stop spamming your song in multiple channels! @humble zinc
You've been warned yesterday too. See our rules #📜┃rules
I feel like Udio is diffusion-based, unlike Suno, which is transformer based. Clues:
- Inpainting (which is very hard to do with transformers)
- Remixing (i.e. changing "style" of a complete piece), very hard too
- Better quality overall
- Harder to change the style once it's there (I'm trying to have metal guitars kick in after a soft intro, no avail)
If i'm correct, it'll be exciting to see the "battle of paradigms"
I've read some papers on DIFFUSION-BASED AUDIO INPAINTING, so could be
it would also explain why ther's no streaming — you can't give "part" of an output generated with diffusion, as it generates the entire output every time (repeating N steps)
with transformers, it goes token by token, so you can start decoding once you have enough (even if not all) tokens
I also love the "choose context" setting with Udio, you can pick how many of previous (or subsequent) settings you want to take into account. This way, you can for example set it to a very low setting (like, 5 seconds), and come up with a natural transition to the next song, even if in a very different style (without having to "rely" on your entire previous song, as it would direct it too much in the same style)
so far it's too "in the rut" though, can't really make it go in an unexpected direction (which is very important in pretty much any music). If it goes on this way, it will be a deal breaker despite all the magical features
if only for some ambient/soundscapes/"relaxing piano" kind of music, I guess
Thanks for giving your take
Real keen to get my hands on whatever infill looks like with Suno
real keen
managed to get it heavy (last two seconds lol), but it took too long (as in literally the song continues too long before getting heavy)
but I did want to write some ambient music anyway so maybe that's what I'll be doing with Udio while continuing to work on metal with Suno
The UI for trimming is very nice. Suno's gonna need to shuffle things round quite a bit I expect otherwise we're gonna have loads of time boxes for everything 😄
yeah, it's perfect, with the waveform and all that
I also love the naming of extensions (2.1.1.2.1), just the way I had it in Jukebox Web UI. I wonder if the creators saw it at some point
Jelly of that waveform select
yeah, guessing where this "0:23" is in your Suno generation is not the most pleasant game of roulette
Since Suno added "extend from" I've wanted to be able to do e.g. milliseconds or at least 1/10th of a second - being able to play the selection in Udio is really useful to hear exactly where the trim will be
also, the sound, it's hard to express. The quality is better, but it's less... lifelike? Let me give an example
I know what you mean
The lyric clarity is great but there is something about the music part
fast forward to 2:20. Like, you hear something like a metalcore guitar sound. It's very close to it. But it's not really what you hear on a metalcore/deathcore record
it might be a good thing, though, coming up with new sounds and genres
Lots of typical music theory can come undone
like, Suno "gets" real music better while Udio tries to mimic it. (Also reminiscent of how diffusion models behave)
It's done "filmscore" pretty well. To the extent I'm going to struggle to add my own musical additions to it.
yeah, probably filmscore is something it gets accustomed to very much as it's very popular. Metalcore, not so much. Just guessing though.
really bouncy, satisfying bass
It's wild that they are so different even though the are the "same" thing
as I suggested, they might be completely different under the hood, transformers in Suno vs diffusion in Udio. If I'm correct.
You can hear it right away but unless you really chew on it like you are you can't articulate what the difference is
It really doesn't generate crisp output though, when it comes to drums or other sharp noises
Besides "music comes out" they seem very different to me
yes, drums are much better on Suno, if you can hear past the artifacts!
Wild idea: Train 2 models, one on sharp sounds (Suno model) such as drums and short string plucks, and another on smooth sounds (Udio model) such as long strings and vocals...
...then combine the output.
and this, again, is a drawback of diffusion models (harder to be attentive to details when you're basically generating an image of a spectrogram)
combine how, that is the question 🙂
Flex tape!
slap
Nah I dunno, I've no idea how any of the AI stuff works beyond a basic introductory level and what I've read here
my verdict: If I had Suno model with Udio UI, I would use it even despite the artifacts (hoping they would improve going forward).
basically we're watching AC vs DC right now (as in Edison vs Tesla, not the Thunderstruck dudes)
I can't get off the feeling I'm listening to "not real" music with Udio. Never felt this with Suno. Then again, it can be a bad thing or a good thing.
I'm amused by some folk treating it like you have to be on one side or the other
I just want waveform select and the vocal clarity
People love tribalism
nah, we can grab popcorn and enjoy
Also inpainting, of course I need that
don't see how that would be possible with Suno tbh, but hey, we can hope
They're going to add inpainting
I hate that neither udio nor suno give you a wav
there's enough artifacts as there are to add more with mp3 compression
Suno devs have said higher bitrate mp3 won't give higher quality (it'd probably give a few higher frequencies) at the moment. I'm sure this'll come in time.
Udio however... 320kbps looks better on paper but hahahaha no it's not that at all.
every small artifact counts. I definitely remember even Jukebox wav's being better (or, rather marginally less awful) than Jukebox mp3's (for the same generations)
True. Suno's main quality issue though is the grainy/crusty/noisy sound, along with the metallic ringing that often develops after some time. I honestly have never noticed the classic mp3 artefacts (back from the days of 128kbps mp3s where it sounds watery/warbly.) on the other hand... Udio sounds like 128kbps mp3 to me, even if I flip between 128kbps and 320kbps mp3s of the same clip side-by-side I hear no difference.
hola
Like Furrybeans said, georg has mentioned it specifically many times. They call it 'infill'
Other devs too no doubt
look up what encodec does. I am 90% suno uses it (like bark did) it is already outputting "compressed" audio. there is no "raw waveform" anywhere in the process
am old man baffled by the clowns on discord - like clown icons clown music... i guess it's just fun for some people?
maybe people like me asking why is playing into the meme
Well, at least the clown memes are much better than the toilet/poop trend we had just before it
yes. its fun. we have a member here who produces pretty interesting content with clowns as a theme and that slowly turned into a insider joke. this creator is inspiring a lot of people to be creative and that sometimes leads to even more circus, clown and tuba content 😄 . i have to say its some kind of refreshing to see some artistically interesting things from time to time, even if they are strange 😄
Interesting to see Udio also gets the occasional "have they changed something? it sounds worse" question
that is helpful context
(though to be fair, they are in beta so it's reasonable to expect things to change...)
Longer version of the absolute insanity I generated with Udio a few nights ago. I particularly found it funny how it vocalised the stuff that was in brackets. I'd say "enjoy this" but that would probably not be the correct phrase to use 😆
WARNING: IT IS QUITE LOUD AND NOT AT ALL MUSICAL
Whaaaat a non perfect udio song must be fake 😏
Hey everyone! 🆘 I need help fine-tuning stable diffusion models for product enhancements. If anyone has experience, I’d greatly appreciate your input.
holaa
I just put the last page of convo here into the lyric box & hit 'generate lyrics' -it made
https://suno.com/song/2e256beb-1493-4439-9ea6-7e538a30c769
oompah circus swing song. Listen and make your own with Suno.
This sounds like a real song to me
I do think it's easier to make a structured song on suno though
Or this
I made a song about my ex gf. I translated the lyrics in englisch for you guys. Just wanna know how you like it.
suno outputs tokens, then there's a variational autoencoder that transforms this into a waveform. I highly doubt the model is configured to transform tokens into mp3 data
@misty kernel , please ping me back plss
@misty kernel
Guys are there any other good Ai like SUno?
you can try udio
Is it easy to use?
I did not say it saves mp3.
Did you look up Encodec? it is a neural compression codec.
It doesn't really matter what happens with the Encodec output.
The point is the model trains on and outputs encodec tokens.
Most of the artifacts are probably from the fact the some of the outputted encodec tokens are not "correct" because they are generated by the suno model and not from a wav file that was fed through the encodec compression. There will be a slight "mismatch" for lack of a better word.
The quality improvement by providing a wav file would be tiny.
But since we don't actually know if suno model works like bark this discussion is a bit pointless anyway. Might very well be it skips the encodec part and uses another tokenization and outputs wav and the devs are lying about the fact the model is the quality bottleneck.
yeah, conspiracy theories aside, I can't think of a valid reason why Suno would deliberately withhold better audio....
But I also can't account for people rushing here to declare Pepsi is better than Coke when it clearly isn't.
Yo @fast rivet I just dm'd you
Well to be 100% fair there could be reasons like "Storage space" or "Bandwidth".
Not that I think they do that. But even then the logical thing would be to make it availabe for extra cost right?
I would... 🤔
(but idk where the cost pitfalls are)
Not much I would think. I mean storage is not free but it has become so cheap
It could also be that 95% of subscriber just want to post a song on social and play on their potato phone...
and the 5% of 'pro-users' wouldn't make up the costs...?
(numbers are made up, obv)
There's a sort of 'mainstreaming' involved in web/ui dev... No point in having pristine files that no one can play.
(but I suspect the reality is AI doesn't make pristine files to begin with)
Is inviting to your own discord server not allowed here? @frank panther , I'm just asking
Suno is not meant to gain you users for your personal server.. it's considered solicitation.
Pepsi is only worth it when it's on sale and cheaper than coke.
the amount of copyright violations on that platform are wild 😄
Hepcat is convinced they are designed to trigger a lawsuit.
hm.. dunno. i mean its a bit tinfoilhat area. i think i reported a dozen and gave up. ^^
Yeah, it is far more likely they are just Silicone Valley Bros and don't consider it a longterm business model.
Not really tinfoil hat territory if you think about it a bit.
We're all living in a bit of suspended time legally when it comes to this stuff. How much human input is needed to make something legally ownable? To what degree is using AI actually infringement?
We have a lot of speculation and a knee-jerk ruling by the copyright office that almost no one thinks is long term.
How do you get a legal precedent? You go to court. How do you get a day in court? You get sued.
There was that thing where "manual" prompt mode disappeared "unintentionally" shortly after someone managed to summon Michael Jackson. No idea what changed afterwards but they do look like they're trying not to get into trouble.
it might be less costly to sue Suno instead of starting a business and basically aim to get sued yourself for damages.
An intentionally bad actor is very much a thing...
I dealt with a guy here, who accidentally posted his "Isn't Suno already over?" script. A list of talking points and how it should be 'framed' in speech.
this was back in October or November. Hepcat would remember it. I saved the script because it was VERY weird.
and VERY specific.
Someone on Reddit called the Suno subscription a "scam" because of how unused credits don't roll over, despite being aware of the message telling you this. They were repeatedly saying elsewhere how Udio is better etc. I smiled yesterday when I saw Udio's handling of credit rollovers (it's the same, but you have to click a FAQ to see that they don't roll over)
I branched my previous weird Udio creation and took it on a different adventure.
tea cup sadnes
Guess this is where I’ll reside from now on
Odd the Offtopic dweller
The Gremlin of Random Chats
It's this one for me. It's always ignorance and stupidity.
Also elevenlabs getting into the music scene, you guys see that?
I'm not able to listen to it but saw someone else post a thing from twitter
I haven't been to Eleven in a loooong time... lemme see what they up to...
Here’s an early preview of ElevenLabs Music.
All of the songs in this thread were generated from a single text prompt with no edits.
Title: It Started to Sing
Style: “Pop pop-rock, country, top charts song.”
goddamn 😄
@frank panther if you didn't see, they linked it
of course they are hand selected - but damn 😄
https://fxtwitter.com/elevenlabsio/status/1788628175766859891
Title: Broke my Heart
Style: “Smooth Contemporary R&B with subtle Electronic elements, featuring a pulsing 104 BPM drum machine beat, filtered synths, lush electric piano, and soaring strings, with an intimate mood.”
Always good to have more competition
And that IS competition.
I wonder if anyone's signed up for a year of Suno, and then a year of Udio, and then...
I am hyped but I will hold my horses until i can hear non cherry picked samples
I mean, there are 20 4 different voices in that song, but it's a song!
And in the end it was all Mr. Tree
This link is meant to inform people on use and proper crediting of Udio's AI.
Please let me know and I will delete the post if you think this is delivering the wrong message. I found it important to know. Read the TOS 😄
🟡Ready to become your own record label? Get started now!:
https://www.topmusicattorney.com/becomeyourownrecordlabel
On this segment of the Top Music Attorney Podcast, Entertainment Attorney Miss Krystle and show/music producer That Orko dive deep into Udio's Terms of Service and discussing the nuances of releasing AI music. They explore the cop...
Thumbnail is click-baity, but information's broken down directly from the most recent Udio's TOS.
Yeah, I like TMA. I think on her stream yesterday, they mentioned they were thinking of doing an analysis of Suno's ToS but prioritized Udio since there were more people asking about it
With all due respect to Suno, and I'm currently paying for Suno's plan, I hope TMA does one for this platform as well.
Anything with generative/transformative AI is important to understand what's required and the inevitable risks and precautions of using these softwares.
Especially since the Government AI-copyright laws at least in USA is not completely sound at this time.
Definitely. It sounded like it as planned to be done eventually by TMA
Not being funny but is there genuinely 40 minutes of content about Udio's TOS? I thought their TOS was pretty clear (credit them, do whatever) and from seeing discussions in this discord server, I understand copyright is a grey area and might vary by region.
(Clickbaity thumbnails suggest brain-rot content to me, so I've not watched any of these.)
It's important to look at all this and be aware and all that but I'm more thinking about the practicality of all this.
Think of how much content is out there already and how much more is being uploaded at any moment
If a song is a song is a song
And you used Udio
and you didn't say you did
How would anyone know
And even if they did, care enough to chase you down to ???? whatever
It just really seems like the legal system and how you engage with it just really can't handle any of this
Especially if it doesn't go mega massive viral
Especially those that remix or redo the music manually in things like FL Studio which makes the audio watermark useless since it will not be present in the rendered version
As far as making covers of songs with Suno. If I add a few paragraphs, but then add a few lyric lines from a song, is that still considered plagerizing and not allowed even though 90% of the song is original
That's when you ENGAGE WITH THE SYSTEM
Since, practically, someone would have to care enough to sue you
and then it would be argued about in the courts for however long
And if you're in different countries then lmao
Possible but the effort required
Backing way up from all this though
There is this sort of taken-for-granted thing that lots of people have with AI stuff, in that they just...say if AI was used then you don't get ownership/copyright of it
But it doesn't work like that
If I, or you, Shadow, remix a thing
Cut it up, arrange it
Alter a part, snip it, clip it, whatever
You 'touch' it
Then that's yours. Or mine. Or theirs. Whoever did it
Then the argument shifts to 'how much did you touch it'
Does that count
But usually the effort required means we never even get there
Covers means you are using the copyrighted lyrics which means you will need a synchronization or mechanical license from the original holder UNLESS it falls under fair use such as with sampling for remixes, etc.
You will need to investigate into what you are planning does not violate copyright (and back up the proof for when it is needed).
I talked about some points regarding this with someone else at #📚┃suno-school message
The point is academic until things go to court
@ashen fjord
I mean, couldn't one also argue that every word is copyrighted since there are a million copy righted songs with almost every word in the english language lol or is it not so much the word but the full sentence? (btw not disagreeing, just curious)
Crediting them is in their ToS, so I'd expect they'd just ban your account if they found out. Or something. Either way, respect the ToS = no repercussions 🤷♂️
Right but your account gets banned is just a you and Suno thing
Totally within their rights
You are still probably fine with using that music
Since the effort on Suno's part would have to warrant them taking care of it
and unless you get all viral
ehhhh
I'm more thinking out loud, mind
I do think it's very wise of them to build up the social angle of Suno
Because that kind of slices out the whole reason people would do those sorts of things in the first place: recognition
popularity
dopamine
While you do make great points for the most part, that doesn't necessarily protect us either KakerMix
It's generally important to be aware of law, and especially how deep into uncharted territories we may be in. That's what protects you against yourself and others. There really isn't an argument you can make that goes against it, however you could try to show others how one might be fine not taking precautions and still being okay without consideration.
Because I like to be involved in the music, I only publish songs with little/no editing that I don't really care about. If I feel "I could really make something of this", then I scrub out most of the evidence that AI was involved by reconstructing the track from scratch. I might just sample bits, so any grey copyright issues would apply to those samples, and good luck if someone wants to break them out of my song to reuse.
I'm not debating against terms or rights or any of that, I agree they are important
It's just the practicality of them
You have to argue in a court of law
If a huge company wants to steal your song they will, you have to chase them down
Totally divorced from AI
Laws are really more of a consequence warning
It doesn't work like that though because the court would obviously reject those trolls (and there is a good chance there are copyright trolls that attempted to do something like that).
As I said before, you will have to look into fair use and other topics related to copyright if you are planning on doing covers and posting it on platforms (and their stance on it).
YouTube definity has guidelines about it which was linked on the copyright removal links in the discussion I had with the other guy.
In the end, it is YOUR responsibility to check out what you are doing falls under fair use or not, etc. as you are the one intentially using portions of someone's copyrighted lyric.
Likely using a sentence is nothing but it would also depend on how unique that line is. If it was made under a parody jest for example, that would be fine due to fair use argument.
The consequence is dependent on the platforms you post to. Suno will likely ban you account if it discovers obvious copyright violation as it is against the ToS (and it had a reminder when you first used the custom lyrics that you must use original lyrics or those you have rights for).
YouTube can end up demonetizing your video/account (given to the original artist), suspension of your account, or even legal challenges, depending on how extent the issue is (this is outright stated and highlighted on their copyright removal page)
Some helpful links:
https://support.google.com/youtube/answer/9783148?hl=en
https://www.youtube.com/howyoutubeworks/policies/copyright/
Fair use is a legal doctrine that says use of copyright-protected material under certain circumstances is allowed without permission from the copyright holder. YouTube gets many requests to remove vi
This is touching a bit on what you were asking about; from the fair use page on YT
Now argue with YouTube about this 
"It isn't the heart" I argue. The person representing YouTube says otherwise.
They police their platform though, and what they say goes
I suppose that's going to be the more important part of the argument, rather than a law one
The law doesn't factor into it vs. how the places where you put the works handle it
But also virality means if something takes off it doesn't really matter if it's stolen or not or who declares it is or not
That was from a list of four factors used to judge about Fair Use in USA (trying to address the question the person I replied to asked about).
As we know, it is ofren more about whose lawyer can be the most convincing rather than proof at times 😭
And the next few years will definitely be interesting
The system is so, so not ready for all this
lol, lmao 
I've said it before but I would be quite worried about long term survival at the level they are at for huge things like Disney
Especially consider how old and out of touch from the business the judges are
Sure Disney owns everything but if the output of everyone is so fast, consumed so quick
and everyone is doing it because it's so easy, what does 'owning' something even mean regarding IP?
I don't exactly think this is a so much near-term issue
But in the future, not too distant, if you can just...spin up shows and movies
Or like with music, right now
You know how much less 'real' music I've listened to lately?
I remember, several years ago, Adobe debuting a demo of their work in progress towards TV and Movie execs for using programs to edit the vocal lines AND mouth/head movement using like a 1920s video (portraying how they can use the program to fix small problems in post rayher than spend thousands to reshoot a scene).
It was so lifelike... and it was way before the whole AI issue itself. I knew then and there would be also a lot of legal challenges about how much likeness a company is allowed to edit and whatnot, especially in terms of compensation if the edits is not small and if they need the actor's agreement.
After all, you know companies would try to shoot and pay an actor like a 10 min scene and use that to make a 2 hour movie with various programs if they can get away with it
I'd think we're going to slingshot right past an actor or a company or a shoot or whatever being needed at all
What is the use for a huge production company
after that, the studio that would need that big production
To me it's pretty clear that for a large portion of people, an AI streaming platform like Suno is 'the way'
At least for music
Shows and movies and long form narratives like that, no
Not yet, maybe not ever
Except for kids shows I suppose, you can get away from quality works needed since the kids would just care about the bright colors and whatnot
Traditional ways aren't stuck, see how traditional animation is just up and gone
That wasn't the field I was going for when I was in college but lmao
Quite a few of the people I went to school with up and had their skillset they built die within the decade
SEE
GEE
EYE
aka CGI, what us cool people were doing
Users just want to jam to some tunes. There are people that care about their artists being 'real' so always a celebrity angle
But other people just want music
and if people make it real, cool. If they use AI but tickles just like the real thing, also cool.
We're all using Suno to make music we want to hear, and the best part is other people who don't care to have any hand at all in creating music listen to music too
If they stream from Udio or Spotify they don't care
Especially once the quality clips over
The painter paints, the maker makes
Get in. We're going to Baskin Robbins.
me, trying to finish my current transcript before I die of exhaustion (I am a transcriptionist)
Drama Queen
@vale lake like the last one:)
thanks you
bruh, I'm having too much fun with these tonight, idk what magic I cooked into that lora but hot diggity
And these are all instant gens, no inpaint
sing along with mike! (that's my real name) - I am continuing with putting a copyright-free version of the Bible to music with Suno. Here's the second installment. The first is above, somewhere... https://suno.com/song/9236a30f-b4a9-4b39-8e0d-4ac6304227d3
may i know how to delete my account permanently?
please write a mail to support@suno.com to get help with internal account settings. ✌️
yeah, elevenlabs music sounds good. i'm much more interested in the ui and control over creations. Udio, even after recent update, can't compete with Suno in stem integration (rap&hiphop). It could easily be with elevenlabs the same, making it comåete with udio for best meme machine, but never with suno for best actual published music
although "broke my heart" from elevenlabs music demoreel does sound like the vocals were correctly stemming with the music. impossible to say if its just luck or actual capability like in Suno, without extensive testing
Udio feels like a half-baked silicone valley cashgrab with their bizarre un-musicality, the paid trolls, and their complete lack of community building.
ElevenLabs is very interesting because their voice model is already great. I subscribed in the past... Want to hear more.
Udio does have some good points. Other than what's been mentioned before, I think the UI is quite nice (at least for the create/extend box and audio range selection.)
Community-wise... they've run a couple of song competitions. Off-topic is discussed in the general channel most of the time. I see Suno being bashed there quite a lot (one person keeps saying it sounds like Windows 95 MIDI - wut?!) One of our old trolls is there, occasionally causing bother. I talk there occasionally and it's interesting to see what changes they make, but I just feel I prefer to chat here most of the time (even whilst I spent 3 solid days playing with Udio)
one should never become a fan of any corporation or product - only quality
it saddens and honestly sickens me to see people and even community managers bashing or blindly defending company or product x or y jus because "it's our team"
if it sucks, it god damn sucks regardless of who made it
I have made a couple of songs with udio myself. I just feel like the vocals lack emotion.
i may be closing in on 30 published songs using Suno, but i'd never start making dumb statements like "suno is perfect" or "lol udio is unnecessary". suno should up itäs game ASAP and include inpaintings and better quality sound, make extension more reliable and let me "mark" voices so I can make duettos with voice 1 and voice 2. Fanboyism kills progress and makes competition just matter of who has the loudest trolls. Its counterproductive, obnoxious and the investors just laugh their beards off
i made exact song yesterday with udio and suno. yes, udios new features were cool and it did sound "nice", but then i went to Suno and after maybe 3 gens I got my mind blown again (which after 20+ published songs itself is odd..) by how well the lyrics meshed with the music and made the overall feeling so much alive and humane
"Product X is DEAD" - when product Y is about to be released
...no, I think what you mean is: "a challenger appears"
don't get me wrong the vocals I made sounded clear with udio, but I just felt like it lacked emotion. Udio does have some nice features, and I do hope they grow in the future.
I hope Eleven Labs includes their 'Style Exaggeration' feature in the music...
I love when AI sounds like flamboyant over-acting....
I think the lack of emotion in Udio comes ultimately from the de-sync of stems I mentioned.
it makes Udio perfect for jazz, but not anything else
glances at Udio discord and raises an eyebrow yep, that's wildly off-topic and possibly offensive switches back to Suno discord
It would be good to see a Varity of choices to choose from
something is wrong if Udio's users prefer to post songs here...
I left the Udio discord cause of the lack of anyone talking. I have yet to post my udio song here tho
They do post songs a lot there (no embeds though, in some ways that's good because less scrolly scroll) but like I said, a lot of off-topic
It's not even comparing them, it's just "Hi, I made a song... over there..."
sunos main is way too fast paced for me too follow. this off-topic is enough for me.
I do that but it's only if it's something really dumb and funny that I think is worth hearing just for it being amusing
I can agree with that. That's why I mostly chat in the noisee discord
in suno I'm mostly like hi listen to my song that I know you will never listen too cause of 10 other songs posted at the same time
i wonder if Suno will also get access to the rumored new chatgpt model. it could make the generated lyrics at least better
and to what extent is the Suno model itself based on gpt-4
could boost it too
yeah. i think this gives a glimpse on what the entire music scene will be in some years. Why would you listen to anyone elses garbage when you like your own
that's why I make videos
that way even if you don't listen to my song if you watch the video it's a win
i might also make some music videos when Sora gets Suno integration
yeah, you should make many videos before Sora and it's 1000 competitiors democratize videos too
good. i published 2 full mastered songs yesterday and they are on spotify today. this will happen to videos too within 1-2 years. Why would you watch other peoples generic videos when you can bring your dreams alive easily yourself?
I like listening to other people's tracks when they are doing something unusual or interesting... It doesn't need to be earth-shattering, just give me a personality and style.
But when it's the audio equivalent of 'lolita anime waifu staring at camera'... I glaze over.
i don't listen other tracks at all. in the beginning i did, but as they seemed to be memes and jokes, i stopped
yeah..., what gets posted in a discord chat..., isn't going to be the sincere stuff...
(unless it's like TOO sincere, lol)
i don't publish anything that I personally don't like. so that naturally leads to me enjoying what I make and listening to it
sounds demanding
v2 for me was that level, v3 already is above that. v4 will exceed anything i believe current finnish rappers will ever be able to do with their average budgets. v5, v6..god only knows
page 3 I think
Just heard the Elevenlabs audio. Is it really that impressive? I feel like Suno does the same, and I can still hear noise by the end of the tracks.
did you hear the demoreels?
they SEEM to be on par if not bit above Udio on vocal and sound quality. Impossible to say if it has as good stem integration of vocals and music and instruments as Suno, but some songs I heard did give that impression
I did hear them and they're impressive at what genre's they're singing.
It's still only hand selected demo stuff available though right?
Not any alpha tester user audio available?
Exactly. We could pick the best from Suno and it'd give just as great quality.
yes, quality/generations is maybe the most important ratio
its godawful in udio for me
basically unusable
yes exactly
I can't use Udio either.
not 100 in suno but maybe like 60-85 often
I get some people swear by Udio, but it's just not for me.
you need to be aiming at publishable music you would listen to yourself actively on dayily basis, not just funny memes for one listen
THEN you will see the wreck of udios current status
Hey I listen to my memes all day okay?
No to be honest I am more in the experimental phase still trying to see what works what doesn't and what does interesting stuff 😄
and we all should be looking for the limits and capabilities. I think the difference is at what do you accept as end result and what are you planning to do with it.
I've been obsessed with my music.
me too, god I listened to my "breakup song" yesterday like an hour straight
when it sounds good already in Suno, you know it will be tear-jerkingly good after some Audacity and BandLab handlings
I am hoping for like Image2image tools that will improve quality of existing songs.
Yea I always try to make my tests as clean as possible. But sometimes it's just not possible and then I just say "eh let's ignore the noise and see where this goes"
I have one or two that could probably be published with some clean up but for now I don't really set that as my goal
Just making stuff I myself enjoy and test stuff out. Doesn't matter if it's slighty imperfect when only I listen to it 😄
i admire your self-control. After like 2 generations I bought the Pro annual sub and then rushed to distrokid and bought the 2nd tier annual sub to it, and started churning out songs
because i want to listen to them from spotify at work
Oh well yea true. I mean I have bought distrokid as well for like 10$/year there was some extra discount on top of their "super special secret discount" (you get when you leave the site and don't buy). And I though eh why not, just in case.
if you have your bangers public in Suno, just publish them via distro
true, might as well
those trending songs are stolen daily for monetizations
This is my issue or I end up spending 4000 credits on song.
Yep, my endings get noise because of this (if it's a mostly test song) because the interesting stuff already happened and I just want it to end gracefully. I still do a few gens but not like extension every few seconds only because the metallic ringing apeared.
My personal next goal is to reach 36 songs, so I can make the maximum lenght album that Distrokid allows, and buy the legacy pack for it. it was 44 euro one time payment.
I'm just uploading to youtube.
Thats the pack that stores it even after you dont pay distrokid anymore right?
yes
Thats quite smart with the big album then. Thats dirt cheap cost/song wise
yeah, i think for one song its 20 euro
so i checked what it was for maximum amount of songs
I actually just started distributing with distro 😅
I gave in, don't know if I made the right choice tho
welcome to the distrofamily
actual bigbrain move
I'm a bit anxious tbh
ahh the first email you get from them "your song is live on spotify and 20 other services"
it was frankly one of the most dystopian moments of my life
honestly just doing the whole process felt so weird for me, kinda surreal haha
and I'm yet to get that email but waiting eagerly 😁
it goes to youtube fast. they will create Artist name - topic
good afternoon
I'm curious, if I already have a channel will it be linked? or maybe I have to do it manually
when you get the email "has been sent to.." it will be on YT within an hour of that. Spotify will take 1-2 working days. Apple Store a week
No they handle everything everywhere
woah 😲 that's great then because there's so many platforms that it'd be such a headache to do it myself lol
how long have you been with distro?
I've never done this sort of thing before lol
since v3 went to aplha
hey zacharia you seem to be knowledgable about distribution. I published my songs to streaming platforms recently via soundcloud for artists. and earlier this week they were all removed for vague reasons citing the music was against their policies. is there some kind of blanket ban on AI generated music that i havent heard about?
it will look like this on spotify
I dont' know about soundcloud. Spotify allows them and so does youtube. To be safe, in disrokid I put my full name in the lyrics and Suno Artificial intelligence (V3) i music
Did your music sound obviously ai ?
so it looks like...
But I guess
far from it, the obvious give away is my artist name is male and most of the song are sung by females
Mine are currently on soundcloud too, so far so good.
hearing you guys experience does ease my anxiety a bit 😊 I do appreciate it
Could also be the genre I guess
I think Spotify is a square image for music tho right?
it's indie-pop/ pop
I’m mainly edm so our vocals are always different
Yes
Soundcloud seems fine for me as well, tho I'm on free so I'll eventually stop posting there sadge
regarding AI rules..yea EU has regulation and individual services have started to take stances, like some metal archive decided not to allow ai music
but so far no bans, only requirements to state if its made using ai
Ai is an instrument of art
omg so cool
it'll be so surreal when I see myself there lmao
can't wait tho
Your songs do well with Suno
But
Idk how well it will do with ppl who don’t expect any ai slip ups in the sound
btw if you have spotify freem you will need...i think 15 songs for a playlist so that spotify won't add random songs. it was my first target
true
You need 30mins total of songs
For an album on Spotify
ah ok its minutes
dang
Some ppl want to put ai in the title for clicks ik
But then you just end up getting criticized more for “it sounds like shit”
Hmm, I see
Bla bla bla
not yet. i have the tier on distro where I can access my artist page, so I made AI statement there to be compliant with EU regulation
I wonder what their stance is on AI music then
It’s like not a fair point of view
yet.. at least in EU there is an upcoming law I think that requires all AI generated contend to be visibly tagged as such.
They won’t score it fairly
I’m not eu so no worries for now
mmm but distro doesn't let you change it after distributing it, right?
Nope
would it affect songs previous to that law?
Will see when we get there
If you hear my songs they don’t sound ai at all
I did use ai lyrics during the beginning so I will give myself that as a giveaway
I emailed Distro about AI tag. they responded with "we don't have that yet, thanks", So i just use the add writer box and add Suno for musioc
okay, will keep this in mind for the next release
I want to be safe... as an EU resident lol
Vocal singers are different usually in every song anyways with edm
or you could also claim autotune
yeah it will take some years until this fuss dies. personally i think its just next step from autotune
yep, I got a year
not paying extras tho
Good
Yeah no need
The hardest part about sharing your music is the beginning
Mobile especially
I finally found a visualizer I like so I will attack that this weekend
I got my cover art album ready too
yeah, tho tbh I didn't even expect to release my music at all anyway
I've used a few for my youtube but honestly I'm not satisfied 😭 but maybe because I such at video editing
like this (assumig your real name would be Chibii Taii)
Same
oh, thank you! I'll save this 🙏
I think my non lyrical tracks will do better this time
btw sick cover art! so cozy ~
Yeah it’s for the lofi playlist I’m almost done with
I need more lofi songs, I'd love to listen when it's out
Problem with creating lofi is you fall asleep
I’ll remember to share it
true haha, either helps concentrate or lulls you to sleep
I think the hard part about being a sumo creator is if your songs are all over the place you won’t have a niche audience
though I do have several playlists for insomnia so it'd be handy anyway
So whoever is planning on making a lot of random shi it’s gonna be tough
I’m staying within the edm niche
mmm maybe not a dedicated one, but you'd have a little of something for any occasion or taste
i'm kinda all over the place ig
but my main niche is just the lyrics I write, nothing else 😅
Once I discovered [louche trombone] my hopes of mainstream music acceptance flew out the window.
Write me some
whost writing
commentary, exciting, tension, uprising, climaxing, spoken words, announcer, increasing song. Listen and make your own with Suno.
rate this beat out of 10
some comparison regarding what i mean with stem integration. Like said, Udio has superior sound quality, but something just feels offy
Live music. Heavy Bass, rhythmic, high-definition, HighDef song. Listen and make your own with Suno.
sunos guitars hitting them bounces is just fucking amazing
the Suno one sounds more like live music... which is the 1st prompt so should be the most important...
30sec on Udio doesn't give it enough time to evolve... it has crowd noise but the mix is a little 'muted'?
udio is a jazzmachine. it shines when it doesn't matter if the band and the singer are not on the same song
i have to admit that the artefacts in udio music really have a way better sound quality.
A summer hit ? https://www.udio.com/songs/oyrVhTM8CJYz1p1hisqxxv lol quite happy with the results, lyrics all made by myself
hmm, the polite frenchman was deleted
maybe he was supposed to post in udio discord
suno off-topic is so udio that lurkers think they are on udio general
udio, is so empty of feedback they mostly here:)
It's almost like they are crap at building a community.
and I'm tired of moderating theirs.
It's the competition, it's like going into Coca-Cola with a Pepsi hahaha 😂
Like, I enjoy this image. But it wasn't what I wanted.
dat table a bit short though
Don't you see the other desks, they're all full!
i have a track with the same name xD
waiting what´s the real result^^
Yeah, it's probably cherrypicked, but they have a great track record with audio so I have high hopes
Can we play with it yet?
As far as I can tell it's not available yet
I don’t believe it until Normies can touch it
.
x3
XD
@uneven shoal sent you a DM regarding the interview stuff. Hope you don't mind me asking a bit more about it.
Well, wow. Just... wow. You just taught me a couple things I didn't know about Suno - and lots more than I care to know about basketball. 😆
It makes sense that ElevenLabs could be totally on-point with the vocals, as they've been best-in-sport on the text-to-speech field for years now. We'll have to see how the rest comes together. I did notice an extensive article on Suno on their site, so they are likely using Suno as inspiration. https://elevenlabs.io/blog/suno-ai/
Suno AI is a tool that's democratizing the music creation process. Learn how the tool works and start creating your own music today.
Uhhh what?
"This article explains how to create AI songs starting from a simple text prompt with the Suno plugin, digs into its top features, and explores the tool's pros and cons."
Suno... plugin?
Maybe someone should shoot em an email with some updates surrounding Suno? They might still have it written around the bots.
The article reads like it was written by AI
have we been had?
That would be hilarious, if it turns out it's just Suno under the hood
"The new Suno killer is.... Suno!"
Hear me out, i remove what's on screen and put Noisee clips on there.
Maybe it's v4 👀
