#🗣️┃chill-chat
1 messages · Page 1 of 1 (latest)
Third I suppose
fourth
pp
The field of generative modeling has witnessed significant advancements in recent years, with researchers striving to create models capable of generating high-quality images. However, these models often need help with image quality and robustness. This research addresses the problem of striking the right balance between producing realistic image...
Anyone else here keep finding themselves staring at the result of a segment, wondering WTF?? only to realize you posted something in the CGPT field and not the other two?
I can't stop making music, what's wrong? 🧩
I think there's a blue chip for that....🔵 🙏
/chirp
Hey @stiff pebble, the /chirp command doesn't work in this channel. Check out the #🐣┃how-to-chirp channel to learn how to make music using the /chirp command, then head over to the #🐣┃chirp-beta-1 channel to make your first song. Enjoy!
Is there any effort in music ML research to generate harmony by chance? I'm curious if anyone is trying to generate close harmony like barbershop quartets.
Let’s have peace soon! 🙏
Music can bring peace together
I've fallen in love with another music model… and it's by ElevenLabs. Out of nowhere Eleven just DROPS this masterpiece and it's like they ripped it right from my dreams, fully formed.
Upload a song and their new AI will turn it into the most legendarily terrible karaoke version - discarding just the right amount of the original performance and pitch data to *perfectly * model a drunk but earnest singer. Perfection. I couldn’t hope to do better than this.
(Dub a video from English to English.)
The dubbing model when used 'as intended' is actually pretty impressive. And when used incorrectly, is actually just the ideal version of one of my Bark projects - figured if Bark was tone deaf, why not leverage it...
😂 🤣 😆
omg..., this can't be real!!!
(i can't breathe)
Лучшие Suno генерации:
Песни на русском.
http://188.226.11.103/suno/index.html
Это уникальный сайт, на котором вы можете насладиться потрясающими музыкальными и песенными генерациями, созданными с помощью искусственного интеллекта как идеи для творчества
best ai applications for making suno outputs sound more mastered? not looking for professional grade mixing, but something to give the tracks a bit more oompf?
I used to use LANDR for my demos: https://www.landr.com/online-audio-mastering/ looks like it's not as free as it used to be 
Honestly, if you're just looking for the basics, then running it through VladG's Limiter No. 6: https://www.tokyodawn.net/vladg-limiter-n6/
and give yourself a little routine that's something like:
- Mess with the knobs for like 20-30 minutes
- GIve your ears a break by doing something else for a few minutes (5-10)
- A/B compare the "mastered" vs "unmastered" in several sections
repeat this like 2-3 times and you're like halfway there
Anyone online for a quick chat about voice cloning?
Interested in cross-language fine-tuning, mainly if I should fine tune (a model like VITS) on the original or the target language
I don't know if anyone knows how best to do cross language Bark voice cloning, it'd be kind of an research project to test out what works best.
I can tell you it kind of works if you just try it in the same language and then give that voice a different language text prompt. Actually let me check one sec
It kind of works if I give a clip of foreign speech (say Chinese) and make it produce my local language (Finnish). But in order to improve the results, should I train on the Chinese speaker's voice+transcription pairs, or would it be better to try to find more data to train on Finnish..?
A basic, generic question, rather than Bark specific 🙂
Well just answering in pure Bark voices. For example, if you give an English bark voice a French prompt, you get a wide range of outputs. Sometimes a total word-salad hallucination. Sometimes you get English accented-French, matching the original voice. And sometimes the audio sample actually starts off in English, or heavily accented, and flips into French halfway. And sometimes even the accent gets smaller or goes away, like, over the course of the audio sample.
So if you 'roll the dice' a lot, you can usually find a lucky sample that basically sounds like the original voice, in the new language. And then since Bark voices are just audio samples (typically anyway) you can use that sample as the new voice from then on.
For Vits I can't really help, just haven't used it myself yet
Are you training bark, or a general audio model question?
More of a general question, Bark cannot generate Finnish :/
Feel like if you want better Finnish, you need to train on more Finnish, but just a gut instinct guess not knowing anything about the model.
You can also just get as close as you can with VITs and use another model to voice-to-voice clone a second time into the target voice, like RVC
Yeah feels like the bottleneck is low text-to-speech quality in Finnish
Just did a quick test to make sure I was remembering correctly about the Bark voice language advice, it'd been awhile since I tried doing it. Seems accurate.
I gave French to Bark with an English voice, and I was expecting to get complete mush, but I was surprised when it sounded like an anglophone making an honest attempt at speaking French.
I know it's actually super impressive really, that it can do voices like that.
It is. I remember as a kid the TTS was only in English, it required some skill to get it to speak some sort of French.
no way 😮
I'm in this picture and I don't like it....
@west hull how tf does this shit work 💀
Go to the #🐣┃chirp-beta-1 channel to use the /chirp command, and check out the #🐣┃how-to-chirp channel for step-by-step instructions for making music using Chirp on Discord.
Aight
I am not into machine learning or coding but I would like to know what kind of knoweldge is needed to make something like Suno or google soundstorm, or this https://x.com/LiuXub/status/1694856130109792587?s=20 I know programming knowledge like python is very well needed and I think machine learning but what else. I asked gpt and it told me neural networks, audio/dataset etc Can anyone help or know more I am interested in it thanks all(BTW I am a non technical dude)
Could anyone ELI5 how https://discord.com/channels/1069381916492562582/1132798201133531136 works...?
Not being a forum police, but https://discord.com/channels/1069381916492562582/1101327266044190721 might be a better place to post this question so the right people see it....
Sorry, I didn't find place where I can post it, thats why I have written here
i made a song about apples
good
how do i make the ai close the song ?
is there a specific prompt , it just keep going and i want him to finish it ?
If it's being stubborn put
Fade Out, Fade Out, Fade Out, Fade Out, Fade Out, Fade Out, Fade Out, Fade Out
or
End, End, End, End, End, End, End, End
in the Genre Prompt with nothing in the lyrics prompt
It's currently broken / stuck on the theme birds, but we're working on fixing it, When it's working, prompts must include the theme or keyword of the day to be accepted / generated — would love to award free credits for submissions with the most likes at some point!
Urban slang for Rap lyrics:
https://urbanthesaurus.org/
can i upload it?
I'm not sure if this should be here or in #💬┃general-chat but since I don't want to spam up that channel I'm going to go for it here.
This thing people are complaining about, when it comes to writing lyrics. That whole AABB thing? Yeah...
I think it's bullcrap. I did some thinking about this, see. Let it get in my head even. Messed me up a bit. But stop yourself and think about it. Who uses those types of rhymes? Why only some of the most successful bands in the world, that's who.
It's good to break rules, I'll be the first to say it. You have to not allow yourself to fall into ruts and repeat stuff too much.
But really, is anyone really going to argue that Journey's stuff is bad because of the rhyming scheme?
Don't stop believin'
Hold on to that feelin'
This is bad because it rhymes?
It's been a mystery
And still they try to see
Why something good can hurt so bad
Caught on a one-way street
The taste of bittersweet
Love will survive somehow, some way
Oh God, even worse, right?
And how much do Eurythmics suck?
My friends know what's in store
I won't be here anymore
I've packed my bags
I've cleaned the floor
Watch me walking
Walking out the door
And that hillbilly, Roy Orbison, what was he thinking?
Pretty woman, won't you pardon me?
Pretty woman, I couldn't help but see
Pretty woman, you look lovely as can be
Are you lonely just like me?
And Whitney Houston? She must have been joking...
I've done alright up 'til now
It's the light of day that shows me how
And when the night falls
The loneliness calls
I could probably pull this all night.
Sure it's a bit tiresome that AI generated lyrics insist on rhyming every. single. line. But it's doing that because that's actually how a lot of songs work. You need that repetition. That alliteration when possible. It makes songs work.
Think of some of your favorite songs and look up the lyrics. How many of them use exactly this rhyme scheme?
So I've resolved to try to stop worrying about this.
Just felt like getting this out of my craw. Thank you for coming to my Ted Talk.
You're WRONG! 🫵
(😓 wait let me explain)
Your examples are biased genre 'Rock-n-Roll' – even bordering on parody of 'Classic Rock Anthem' (big energy, simple structure, loud guitars). Yes, if you want the AI to make that sound, follow that song structure.
The 'sophisticated rhyme' debate is more about rap, which evolved over decades – so people should be sharing that info, it's cool. If you work a classic rock style with an interesting rhyme scheme that creates a new rhythm..., good. That's some crossover.
but yeah, no hate on nursery rhymes, they are structured that way to have a hook. It works. Kids love it.
So the problem I see is the insistence on a rhyme (at the end of every line) at expense of a coherent sentence (mangling normal sentence structure – let's admit ChatGPT is not writing Shelley)
... and wasting lines on random idioms JUST for the rhyme, not because it fits the tone or the subject....
In your above examples, only "My friends know what's in store" is a throwaway idiom, just about everything else stays on topic, each line contributing... while rhyming. The rhyme is just icing, It's not cake.
This is an AI fail, it's making something that looks like a poem with no understanding of poetry. People are asking for Camelot, they are getting Castle Burger. Sure there is an endorphin bubble in expecting a rhyme and getting one..., on par with playing peek-a-boo. It's impressive, but shiny AI newness is wearing off..... Humans can always go to the next step just by making an effort (just edit a few lines, it will be better)
BUT the practical question here might be: how does it effect the song structure?
What if we can control the rhythm and pacing of the song, with syllables and rhyme-structure (it appears so based on my songs using the Sumerian transliteration... separating the same lines in different ways triggered different effects on the music)
Yeah, rhymes are great, but don't get stuck in a rut where you can't do anything else....
So it's a lot less about the fact a rhyme exists in and of itself and more about the forced nature of it when the AI contorts to make it work? As someone previously pointed out, Mr T's infamous "Mother, there is no other" rhyme?
And it's not a reflection on anyone's lyrics unless they too are engaging in uncomfortable contrortions of that nature?
I don't think it's a controversy to say:
"Mother, there is no other...."
is more interesting (aggressive rhythmically) rhyme than:
"I'm a Barbie girl, in a Barbie world...."
But if my goal is a silly pop-music sing-along with an insipid hook designed to stick in everyone's head, the 2nd lyric nails it.
Pop sensibility is often about lowest denominator and a hook. "Pretty Woman" is the hook that everyone sings, it doesn't matter if "Won't you pardon me?" rhymes or not. That line is disposable (I literally didn't know if that line was accurate, despite hearing the song countless times).
The problem is debating about rhyme-or-not is a false dichotomy, not that helpful when the actual marketable parts of the song aren't mentioned at all.
It was only ever meant to be inspirational advice for other people writing rap lyrics.
I'm sure we've all made fun of Madonna for having a 6-rhyme vocabulary, but she did ok financially.
I don't remember anyone saying 'STOP USING RHYMES' altogether .... That seems like a strawman.
LOL, well it wasn't entirely here that sentiment showed up. I did some searches about this before posting above. People were really laying into writing song lyrics with rhymes. Like really laying into it.
And at first I was nodding along because yeah, it's a bit attrocious and sometimes unnatural. When I have the temptation to slot in a word that deserves an SAT score I stop and ask myself who I am trying to impress here? Because what good is the rhyme if it is to a word nobody knows without a dictionary? If you're so desperate to have a rhyme, you may as well use a made up word, it works well enough...
But then I started really thinking about it, and I realized that a lot of the songs I knew and liked had quite a bit of rhyming in them.
Oh I know that I cheated above, a lot of the ones I cherry picked were prechoruses and other parts which indeed use a rhyme as part of the hook.
But I really thought--hey, wait a minute here--I'm throwing out the baby with the bathwater!
Now if there is a specific discussion that was had here that was about rap only, then I apologize for coming at it from a different angle, I just saw a lot of discussions about it when I went looking and I'm addressing the whole and not a specific genre.
Ahh.... yeah. I'm sure there are parts of the internet that are advocating terrible extremist advice.
🫡 Carry On, Soldier!
😁
See, I'm not trying to start a fight. I just thought this might be a discussion worth having.
😋 I was happy to be a guest on your talk show...
So just so everyone know, I haven't checked out today (though my nose probably wishes I had) I'm actually trying to use Moon Valley to generate a music video...five seconds at a time.
So that's why I'm a bit distracted.
Ha, I love this discussion. I definitely think there's a time and place for clever rhyming across genres. But seeing GPT always do the same AABB quickly gives you an appreciation for well crafted lyrics 🙂
Oh I don't disagree. ChatGPT does it too much. Definitely.
But the way various discussions were going seemed to imply that it was always bad.
I don't know that it is. I think the question is how awkaward do the words sound when you try to say them yourself? If it's not something that you can nod along with and mouth the lyrics to (I really can't and shouldn't sing) then it's probably not a good line.
I have no problems with using ChatGPT for ideas, it's a decent writing partner when you're stuck. But I rarely use its exact output because of precisely the issues mentioned.
Hey Ryan, you can use the #⏩┃continuation feature. Check out that channel for more detials.
thank you so much
hello!
Well that's Bark isn't it? And some other recent models. Well I guess I'm not sure what "acoustic" means, just read that as "audio"
Or you mean pure audio maybe, audio in, audio out
I am trying to build an arabic TTS Model or atleast find one..
i also found this https://github.com/ARBML/klaam
this is the model https://github.com/tttzof351/SimpleTransfromerTTS
the 2nd model is described by https://medium.com/@tttzof351/build-text-to-speech-from-scratch-part-1-ba8b313a504f @grim idol
The first model on the other hand uses a method called fastspeech2
Well finding arabic tts is way way easier than building one yourself. Are you trying to build your own TTS for a project, or do you just want to find one that works? xtts works on arabic https://coqui.ai/blog/tts/open_xtts
I can't help you build your own TTS from scratch, that's on my list to learn soon too.
how to increase the song time?
instructions to https://discord.com/channels/1069381916492562582/1149508588470022164 a song after the 1st chirp
Referencing the #🗳┃mini-polls most current as of this posting, I'm interested but only in complete songs. And not all the time, just sometimes I'd love to be able to just hit listen and then if something hits me hard, be able to go back and see what exactly it is that they did for inspirational purposes.
But right now most of the ❤️ 'songs' aren't even that.... they're just pieces. And the English language stream is usually full of every language but English.
So probably the best thing would be some kind of curated collection of stuff with a way to link to the posted and publically available video of the song, so interesting ones can be looked at in context, as it were.
Great feedback!
Suno is so censored
There's a whole world of singing AI out there I barely knew existed. (In part because it's mostly non English) This converts AI singer projects from one format/synthesizer to another https://openvpi.github.io/
I was intrigued by "tuner" as the Chinese word for people really into AI singing. According to ChatGPT it doesn't sound as cool in the original language though
Yeah, some of these interfaces are incredible.
Bot music analysis is a scam 🙄
🤖 yer genre is... [looks up 10 most popular genres according to the internet]
Rap-Rock-Pop-R&B-EDM-Latin-KPop-Country
Ha, what's your self-described top 3? Curious
idk i waz tryin to find out 😭
according to Chirp (lol) my favorite genre is "Funk Bass with Compextro Tribal Percussion"
😹 I need to find a shorter name for this genre...
Not really...a subset of pop I'd say 😄
Its pop
But korean
And toxic
СТИМ ДАРИТ ПОДАРОЧНУЮ КАРТУ НА 50$
https://urlis.net/card-games
K-Drama I understand as a genre form watching an way too much Netflix, but has distinct style and conventions. Presumably k-pop does but I have no idea, I just think of k-pop as the name on a spice in my cabinet of Chirp prompts. "Add three shakes of k-pop..."
there are best tool deepfake like roop?
not here.
😆 See? It totally worked!
I got the pro version thinking I can extend to 2 min. Anyone can help?
Are you continuing when you create segments?
some channels aren't showing to everyone...
here's the page with the instructions for https://discord.com/channels/1069381916492562582/1149508588470022164
stands on soapbox
I would rather chirp music than play videogames or watch netflix.
It might just be killing time, but at least with generative AI I have a souvenir at the end.
🎵
tfw re-chirping 8 times to get the right voice for the song....
https://www.youtube.com/watch?v=0RGtgZJCc-o
Taken from: Phantom of the Paradise (1974)
Written and directed by Brian De Palma
/chirp
so is this a scam?
Hmm, probably. Never even heard of 11:11.
more like...
True
Can I use purchased songs to register for Network?
question for people that use any websites for sharing AI art: what features would you like to see?
В чем прикол? Мне пишет эту штуку "This prompt has been flagged. Try again with a different prompt." и ниче не происходит
Feeling pretty good today.
Recent changes in what was available for use in the How is it Manifested image generations have resulted in gaining higher quality images, which allow the Gamer channel to start looking like actual video game screens for fictional games, even better a revision of the channel prompt has resulted in some serious improvements in the structure of the created prompts, if I do say so myself. And the work done with the Cartoon Show channel prompt is outstanding.
So I encourage everyone who is into AI and AI generated television to drop in tonight at 8:30 PM CST and give it a look.
https://www.twitch.tv/howisitmanifested
That's when live generationn of episodes begins, and continues on until midnight on weekends, until 11:00 PM CST on weekdays. All other hours we play reruns. I'm also a moderator over there too so if you visit that Discord you'll probably run into me sooner or later.
1082792484700180531
Anyone else think we should be allowed to have swears in suno bot
I'd be happy just not having my song flagged for the word "stupid" in it.
There's that of course, if I were just going for a random insult that meant person of low intelligence. The problem is that I need a word that means that and which rhymes with the scheme I am using for my verses, which is a bit harder.
(Or were you talking to the other guy?)
Anyway, sorry for being pissy about things.
Just frustrating.
New OpenAI TTS voices quite odd. tags are clearly seen by the model, adding more adds more time, sometimes you get a perfect laugh, other times just this weird stuff
Using openai TTS also for our live speaking app. https://ai-cartel.com
AI Recruiter is an artificial intelligence hiring solution that automatically finds, calls and interviews top candidates without human interaction.
/chirp
Go to the #🐣┃chirp-beta-1 channel to use the /chirp command, and check out the #🐣┃how-to-chirp channel for step-by-step instructions for making music using Chirp on Discord.
sus or local server feature? @warm laurel
Go to the #🐣┃chirp-beta-1 channel to use the /chirp command, and check out the #🐣┃how-to-chirp channel for step-by-step instructions for making music using Chirp on Discord.
Are charges for credits recurring?
type /subscribe
to adjust your subscription on the Suno website.
hello
I made custom GPTs in ChatGPT:
https://chat.openai.com/g/g-gnsbwpPxu-code-helper-buddy
https://chat.openai.com/g/g-DVW0bHuCa-iconify
|https://chat.openai.com/g/g-sd67rEBZC-character-creator-gpt
Thank you
crea
Even if Epic Games shut down their servers: Unreal Tournament can still be played online!! Go go go :)
Fun for each and everyone! Beginners as well as veterans are welcome to join the Friday Frag Nights!
1 server to rule them all: DM, TDM, CTF, Domination, Assault, MonsterHunt, SLV, UTRoyale, GunGames, InstaGib, ComboGib, Sniper, Co-op games, etc. etc.
After each round the players can vote for various mods and mutators - including tons of maps...
how can I make the songs chirp generates sound like what I imagen them?
WIth lots and lots of practice.
what do you mean? I only see options for telling it what genra and the lyrics.
Make sure this one is on Chirp and not Bark:
https://suno-ai.notion.site/Using-Metatags-34944efe09ec41d693e314f13af44695
Lots of tips here:
https://suno-ai.notion.site/Tips-Tricks-Chirp-v1-969852c74c644c6b8262ec5c5be2325c
Metatags guide prompts but aren’t verbalized themselves. Like stage directions or parentheticals in screenplays, you can use metatags to describe a specific part of your prompt (e.g. [laughter] and piano interlude) or to influence your entire prompt by describing speaker characteristics like [whispering] or music genres like rock.
Brain read this as With lots and lots of credits lol
lulz
Well those can help....
Casually wonders if and when there would be the option to apply for moderator
apply for deez nuts
OHHHHHH
DM @amber hemlock
👍
One heck of a shake-up at OpenAI tonight...
i think the weirdest part is all the AI bros crying for Sam Altman, saying he's a victim and is scapegoated.... 😞
but if someone is a scapegoat, uh, shouldn't the reason be waved like a flag...? I suppose there's still time, but what is the supposed reason... which has already been deemed false before it is known?
😩 weird compulsion to worship and protect wealthy white men they don't know.
whole thing seems really strange, I don't know much about the current company but used to chat (well twitter replies) with a few folks very casually when I did weird GPT-2 stuff, but I think every one of them is gone now
I feel like this isn't off topic enough for a Friday night so I'll also add, it's a myth that roosters only make noise in the morning and I wish i didn't find that out the hard way (from my neighbor having one)
yo
So whats all the OpenAI stuff about? What happened?
Anyone from Suno can fix my Discord subscription? It's linked to my google account, but I upgraded and it's still stuck on free plan
email billing@suno.ai
Sam Altman is another person the media loves to focus on because he's very wealthy and says crazy stuff sometimes.
(we could stop there)
The board of Directors at Open AI just fired Sam as CEO and voted him off the island for making false reports to them ("not being candid"). This is unusual because normally a CEO is half paid in company stock, so the CEO has an invested reason to do whatever is best for the stock (if not the company). And because the CEO is (normally) a huge stockholder he can vote himself a golden parachute (basically cash out on exit).
Sam had weirdly NO Open AI stock, so the board had no problem voting him out – which raises other questions like, where were his loyalties – was he a Microsoft plant all along, was he double-dealing, what was he lying to them about....
So everyone is salivating for the juicy details which will probably leak 'anonymously' any minute.
Okay so thats funn..
I suppose we can ignore the Less Wrong article about the guy's sister and her allegations of multiple thing including a protracted campaign of shadowbanning across platforms too....
Yes, that Less Wrong.
I wish Midjourney would give us access to the older models for free
And allow us to pay for the newwr ones
V3 was the best in my opinion
So surreal and dreamy
@bronze rock happy birthday
😒
however good your new TTS is, do everything in your power to avoid having the very first sample on your demo page be LJSpeech lady. i just can't
I see nobody mentioned StyleTTS 2 here
I dunno what happened but somehow it's under people's radar now, altho the git repo has been there at least one month
ah, found ref here #💬┃general-chat message
I just looked at it quick, that's actually what my sentence right above yours was about lol
yeah I kinda wondered if it was about it
I dunno why the internet suddenly realized this exists
I'm bit puzzled about it
🫗
Not as good as Xtts, neither as good Bark really in terms of the stuff I personally like (expressivity, surprise), but seems to be actually free unlike XTTS, the 'style' part is super cool, and inferencing is super super fast so maybe ideal for a chatting app or something. Install stuff https://llm-tracker.info/books/howto-guides/page/styletts-2
my issue with bark is not really the expressiveness - I love that part, but it's really hard to make bark sound consistent
Yeah, I get that. Up until recently I was almost always in exploring or playing mindset with TTS, not deploying products, not even producing an specific output with a goal, often not even something shared. Just: let's see what happens or what I can make happen. If it took 10 gens to get one funny silly-accent-Obama clip I'm happy, the only metric was making myself laugh.
My Way of Life cover by Sheldon Plankton https://youtu.be/_xt9XFwe2cw
HIGHER OCTAVE VERSION: https://youtu.be/hSdzWpJNn28
My Way of Life by Franksinatra
Hey @umbral wadi What AI are you using to make music video esk stuf trying to find one that can do some sort of Image to Gif generation
I've tried a few. The ones I have used are:
- Moonvalley.ai
- Pixverse.ai
- Morphstudio.xyz
- Runwayml.com
- Pikalabs.org
The ones I have actually made a video with were Moonvalley.ai and Pixverse.ai
Music video for the Praise Stream song, "It's Your Goodness" all footage was created using MoonValley AI in five second increments using their open beta to produce. Very nice tool and even nicer developers over there. Well worth checking them out!
Moonvalley:
website: https://moonvalley.ai
discord: https://discord.gg/moonvalley
#praiseandwors...
Newcomers Poppy Felicite take you out to the beach in this new song. Nice beats, and a very addictive bit of music there that will make you want to do a bit of dancing yourself. Much thanks to PixVerse for their text to video AI which produced all the video seen here.
Ocean sounds from https://freesound.org/people/OldPueblo/sounds/353875/
Sea...
👍 thanks!
No problem. I will point out that just because I haven't made a video with something yet doesn't imply it isn't good.
Oh and I forgot to mention, I use Kaiber.ai for my audio teaser music videos.
Well obvious also runways pretty good :]
It can be for somethings. I just seem pants at using it for what I want.
Ah i thought that was real
I knew it would fool people soI grabbed the link when I saw it.
Well that however is real
Like no joke is real
and were back
Jeff's getting overwhelmed.
I tried this a bit more, I just can't anything out of StyleTTS 2 that sounds as good as these clips on their page here. No matter what I do it comes out a bit too Spock-like, over enunciated.
idk if I'm allowed to post this here, but have been building https://karaokesmash.com with some friends and figured y'all might enjoy it ❤️
Damn..
I wish they'd succeeded in removing Sam Altman
I'll take a look and see if we can get it added!
one doesn't simply UNDO a pact with the devil...
Clearly. But I'd say Altman is Baalzebub, and Gates is Shaitan.
It's surely MS that blocked the change. But it would've been nice if OpenAI had been able to go back to their actual mission, which they've abandoned for years
it didn't quite get everything right, but my prompt wasn't really THAT detailed... https://karaokesmash.com/app?id=3qWgUV5wJsfvkTBJZ09d
What do y'all rate this fellow
where do I find the stuff I hearted on the website?
hard to find my old clips I have so many haha
nm its at the top favorites only
Can someone tell me more about the uses allowed with the generations made by V2 on the website, if paying for a subscription?
with free subscription the license is "in-house" non-commercial, but we can share the song on (free) social media with attribution to Suno (video shows their logo, so that counts)
with paid subscription, your chirps get a "commercial license" which allows you to monetize (spotify, apple music) and use in professional projects (animations, advertising, film)
the license is perpetual, it 'sticks' with the chirp, so your subscription can change but that chirp is still under the license when it was created.
Wow that is pretty cool, with the commercial license we can only monetize through streaming and not really "own" the songs right? Like no selling to others. Also what if there are copyright issues on social media with any generations? Or is this not possible?
so this part gets a little complicated, and is (confusingly) explained in the TOS on Suno website:
Discord is social, so I can build on your public chirps, and anyone can build on mine.... I can't be sure if your lyrics are original or copied from a current popsong... it's a copyright tangle...
so when we submit our prompts, it all becomes 'owned' by Suno, and they grant us back the perpetual licenses. As the 'rights holder' Suno is allowed to 'distribute' the music (host on servers, play on FB...) even if we monetize.
Music industry is a powerful lobby, and they have created elaborate licensing and distribution laws, because copyright wasn't enough (even tho they changed those too).
the licensing thing cleans that up...
If you are making a song for a client... 🤔 ... you might want to look at the TOS yourself, or have a business law advisor take a look... (don't rely on my layman's explanation)
Uhhh I'll just stick to streaming and monetizing that way and hope I have no issues ?
yeah, the license is forever, so Suno can't charge you a million dollars in shakedown money after you have a hit... lol
Okay that is good to know. Thanks for the info. Just one more thing, any advice how I can use it to make money? I have ideas but just not sure where to start. I have a whole bunch of other things I'm working on already. Not sure if I want to include what is generated from this AI into what I have now or not 🤔
😬 not a lot of people make a living on their music alone....
but many members here are using Spotify and distrokid (et al)...
it's probably too soon to report any income, but some report a lot of listens.
(AI get-rich-quick schemes haven't really panned out for creators, afaict)
Yeah well it seems they keep making it more difficult to get tracks on Spotify, at least from what I've seen.
How not though? I'm sure some people have done it. Youtube definitely tells another story lol
I think the goal posts will always be moving...
And music is not AI art... no one can stare at ai art all day...
but music... is a lifestyle.
(I think the music industry should be worried)
Yes I am worried and I use AI lol Especially after discovering how good Suno has made their generations. 😬 Before discovering this world I was learning to make beats and music in a DAW all myself but ever since AI has advanced so much I have gotten lazy lol If I can use it to help me get some money coming in thought I definitely would take more time trying to learn skills that will keep my from being obsolete when AI takes over lol
Turkey Time is just beautiful
So what I created before starting the subscription does not have the commerical license right?
My concern is that I like my own stuff so much I begin to have a hard time listening to anything else after a while.
So what happens if this is the way of the future, and we're all just listening to our own stuff? That makes it a bit harder for anyone trying to make music in the industry, doesn't it? ...including any of us.
Sure there will always be those who get better results than others or have extraordinary luck in their generations. Sure there will be people who differentiate with music videos, spectrum audio graphs, etc...
And there are always going to be those whose sheer talent and experience puts anything that can be created by AI to shame. (Sorry Suno, you know I love you guys but it's true, at least for now.)
But if we're all just listening to our own personal radio made up of our own generations...what then?
Sounds ideal actually.
Creative entertainment without a cult of ego...?
👍
I get what you're saying but I just want to play too, not take the bat and ball and go home with it.
Good music without ego shenanigans (AKA sex appeal) is obviously how a record label goes broke... 😛
Good point.
Meanwhile I continue to beat my head against the brick wall that is my version of Nights of Splendor....
So why can't we listen to our music and shared it with the world at the same time? 🤔
No one is saying we can't.
because (hypothetically) everyone is doing that...
The implication is we're too busy listening to our own stuff to 'bother' listening to anyone else's stuff...
therefore no one listens to each other's music, only their own
Oh okay I get what you're saying now lol
We can always hire attractive people to jump around on stage while the music plays....
Argh...lost my voice again...
I really wish I could start at the first one, trunicate before the instrumental broke my chorus and tell it to keep going...
Maybe in the next version?
Giving one of my other whales a look. Maybe I can finally fix that bad sound shift for In the Middle of the Night?
I'm sure we've all got some of these. Songs we were really impressed with and then fell off. Here's a second chance for those, right? 😅
🥺
We don't have to have to give attribution to Suno when uploading do we?
Depends on the account you have.
not with a paid subscription.
(only with free subscription)
Remember though, even if you only have a free subscription as long as you're sharing the intact video it already has Suno logos making the attribution clear.
I'm already on a subscription lol I jumped on it fast
Then if you're on a paid subscription you're fine.
happy t day if youre into that
Guys, how can i get coupon code for a discount for subscribe suno
protect the leopard ❤️
https://youtu.be/kVc0vufYKxA
Mani Flo - Afro Hype maintenant disponible
sur toutes les plateformes
https://bfan.link/mani-flo-afro-hype
Follow Mani Flo :
Facebook : @maniflo
Instagram : @maniflo_cfm
Tik Tok : @maniflo_cfm
Contact Booking :
prodmorelevel@gmail.com
Réalisation vidéo :
@covergrafik
Produits par Cainfree Music
#maniflo #afrohype #musiqueafro
How long does it usually take to get a response from the support email address?
Please bear in mind that this is considered a holiday weekend in the United States. I am sure you will get a response as soon as it is possible for someone to send one.
You may be interested in #1151542418991239168
I got to say, yesterday was the first day using Suno.ai on the website and I am just blown away. One extra detail I think is amazing is it does Spanish lyrics just as good as English. Idk if it does other languages as well but I just know it says Spanish just as correctly. My gf was even very impressed with it and she is a Native Spanish speaker and not even into AI like I am lol It seems to be rare with a lot of generative AI I've ran across to create in Spanish or other languages good. Great job Suno, definitely going to continue following this AI.
wait, go ahead and say the context here...
😅
this is from Stable Audio... or...
Yeah it's Stable
I actually expected more of an "actively racing" sound than a "driving onto track" sound
Still good tho
I put in this prompt instead and got a better sound from it
This is great! 😎
I need to check out what's going on over there...
Haven't tried since they launched...
meditation music
ello
/chirp
Go to the #🐣┃chirp-beta-1 channel to use the /chirp command, and check out the #🐣┃how-to-chirp channel for step-by-step instructions for making music using Chirp on Discord.
How can I generate MEDITATION MUSIC ?
try words in the style prompts like:
ambient, ethereal, slow, cinematic, call to prayer....
I've been trying to get a red-army choir type of voice, but could never manage, is it even possible?
THANKS, i gonna try
I haven't had much luck prompting choral voices... (seems to work better on the chorus – coincidence?)
Here's an attempt with 'trio' – which sometimes works...
Guten morgen kann ich auch lieder in länge von 4min machen lassen? mega cool..!
Verwenden Sie diese Taste, um weitere 60 Sekunden fortzufahren
(use this button to CONTINUE another 60sec)
I've been experimenting with an AI radio where the songs are generated via Suno.ai. It's still a work in progress, but what is your guys thoughts on this?
/chirp
Go to the #🐣┃chirp-beta-1 channel to use the /chirp command, and check out the #🐣┃how-to-chirp channel for step-by-step instructions for making music using Chirp on Discord.
Aprende a CREAR MUSICA CON IA con LA MEJOR HERRAMIENTA hasta el momento: SUNO 🎶🤖 https://youtu.be/Tsfh5Xyt_1U
¡Bienvenido a una emocionante exploración de Suno, una herramienta de inteligencia artificial generativa que está revolucionando la forma de crear música! En este video, te llevamos a un viaje a través de las increíbles capacidades de Suno, mostrándote cómo puedes dar vida a tus ideas musicales con facilidad y creatividad.
🔍 Lo que Aprenderás 🔍...
Can we use artist voices?
By design this is not allowed.
/chrip
Thanks , it was my doubt
Yes, however, it a bit of a complex system that utilizes lambda functions, etc. With sensitive information on the private repo which I unfortunately can't make public 😦
Not sure, but I don't think an api would make sense, because one single good song could take around 20 chirps in order to get the tune right. So genrating it via an API wouldn't really make sense.
No, the songs itself is pre-made song I generated via Suno
You mean like the channel #🐣┃chirp-beta-1 ?
Hey guys, does anyone here create custom GPTs?
I found an incredibly powerful workflow that I'd like to share, but not sure if there is a more proper discord or subchannel...
Turns out, you can program your GPT to do just about anything!
https://www.youtube.com/watch?v=kGsxhwmCp8A
And they have a free program at the bottom of their store page for 100% discount on premium products too!
Introducing GPT Infinite: Empowering Your GPTs with Infinite Possibilities.
In the rapidly evolving landscape of artificial intelligence, GPT Infinite stands as a trailblazing solution, well ahead of the curve. Our platform offers a comprehensive solution to address GPT creators key issues.
Issue one, Enforcing Instruction Compliance,
Solution...
Hi Have you read rule #4️⃣ yet?
@umbral wadi Hey, sorry, I didn't think about it like that... Is it considered advertising if I share details with the community about cool new tools for ai?
At this point of time, pending a different ruling by team members I'm thinking it is. Especially with talk about discounts and most damningly the double posting in more than one channel.
So for now I would appreciate if you would put a hold on this until I can get a ruling on it.
Youtube videos I believe are fine, but once you post links to random sites, that might be legit but most of them are scams and therefore it just looks shady. Imagine if someone posted links + discounts with limited number of people to enter on a random site you never seen before, you would sus it right?
My god, the video has 18 views and 1 subscriber, are you sure you haven't made it yourself or uploaded someone's video on your channel?
You also immediately changed your name and profile pic, there are many reasons you would do so, but everything leads up to you acting quite shady
I gave you the benefit of the doubt, but I wouldn't trust anything you post
I changed my profile picture to an actual picture of me, so people would 'NOT' think I was some spam bot. However, I do understand double posting is spammish. I will be more careful about that in the future, thanks for the warnings.
Then delete all of your posts about the website and the youtube video you created
I love making songs about my cat!
Is there any security issues with linking audio/video? There isn't a good "share" but it seems others can access the link when it goes to download, but I don't understand what is involved in the url
ie https://cdn1.suno.ai/<stuff>.mp4
How is that <stuff> generated I guess?
Unique share link (similar to how youtube/google docs etc does it)
Are there rules against commercializing a song we created through Suno?
Like to know the same question John asked - Are they any rules against commercializing a song we created through Suno???
hi
@zinc sundial , @fossil karma
paid subscription gets a commercial license on all generated chirps.
free subscription is non-commercial 'in house' license (ok to post on un-monetized social media with attribution to Suno)
Hi everybody,
Can I ask you something
How to create a female or male voice, high or low voice as you like?
You can try [male high vocals] [female low vocals] or [female, low vocals] and such. But keep in mind these metatags are not official and will probably not work most of the time....yet, sometimes bruteforcing like that works
Thanks @frank panther - how about stems? Better quality? In the roadmap?
some 'edit' features have been discussed, and the ability to start a continuation from any point in the song... Also different features costing more or less tokens...
(I haven't seen an alpha or anything, so I assume it's in the R&D phase...)
not an official roadmap, but Suno has mentioned upgrading Discord to v2, and some QoL improvements on the website to manage our songs... (so I am estimating any big new features to be probably in the new year...)
https://youtu.be/97iMxC3FF6E
I like this guy's channel.
He usually reviews / interviews music related topics.
In this video he reviews some other (inferior to Suno) AI music service.
People in the comments section are telling him to check out Suno next. 🤘
He has 3.79M subscribers... if he does come here... the userbase might explode. 🤯
In today's episode, we listen to some generative AI music in different genres.
📚 The Black Friday Bundle Sale — $89 FOR ALL OF My Courses: ⇢ https://beatoblackfriday.com/
📘— The Beato Book Interactive - $99.00 value
🎸 — Beato Beginner Guitar - $159.00 value
👂— The Beato Ear Training Program - $99.00 value
🎸— The Quick Lessons Pro Guitar Course...
cc @amber hemlock
Okay so two questions, one: Is there a way to prompt to make it to where the song does not sing about the genre or style of music we are prompting for? Two: Is there a place we can share our creations?
use the Create(Custom) link – 2-prompt ui
lyrics and genre are separate, and a new button to Generate Lyrics before you generate the song (edit the words, re-roll, write your own)
Oh okay thanks
Okay now one more... How do you make a longer song that stays consistent?
when you Continue From This Clip, it listens to the final 20sec or so of the clip, and continues from there in a new chirp, edit the lyrics, if any words got cut off pick up where it left off and continue the song.
as long as the voice and some of the chorus is heard at the end of the clip, you should be able to continue it, more or less consistently, and repeat the chorus in the next clip..., but it might take a few re-tries to get something you want since there is 🎲
Just saw an article about this: https://pika.art/ Imagine a future Suno+Pika tool. That would be so cool.
Sorry, another image post-related post, but this tech is just crazy. https://clipdrop.co/stable-diffusion-turbo Changes the images WHILE you are typing. Are you kidding me! Wow.
Clipdrop - SDXL Turbo
i hope we can get close to this with audio one day 🙂
The quality drop tho, reminds me of dalle-mini. Besides it said: "You have reached the usage cap". But why would anyone pay for a subscription for that?
I do believe we will reach a level that anyone in their home computers will be able to generate high quality stuff with AI without tons of GPU power like we need today. So imagine Dalle3, ChatGPT, SunoAI in your home computers. That would be insane
a (fake for this film) Music AI in 1984 on a 8088 PC with 1024 Kbyte RAM (lol) 😉
https://youtu.be/fc8yjOrGUhQ?si=wmB-zPjgplXxNWIN
Nice find!
Back in my undergrad I had some idea about "deontic logic + automated theorem proving", and that was before the ML explosion, what a fascinating time, wonder if/when ML will get involved in law, pure PKD material haha 😆
this is fire
One of them was me 😆
That would be incredible if Rick joined us
He's in for a treat. I think he will be shocked
Maybe people could prepare a hard rock with a guitar solo for him to hear
Watch the full show here: https://3voor12.vpro.nl/media/collecties/video/festivals/lowlands/2019.html
LL19: Giorgio Moroders uitzinnige meezinguur
Geweldige revanche voor mislukte dj-flirt https://3voor12.vpro.nl/update~531cd422-1153-4b6e-bd59-4fe434693c87~ll19-giorgio-moroders-uitzinnige-meezinguur~.html
Setlist:
I Feel Love
Hot Stuff
Call Me...
Alright. REAL TIME?
Still needs to be baked a bit more. https://i.gyazo.com/c0acc483b65ef9196a1c16f0db7e85b6.png
Provided to YouTube by Mute
Metal Fingers in My Body · Add N To (X)
Avant Hard
℗ 1999 The copyright in this sound recording is owned by Mute Records Ltd
Composer: Add N To x
Auto-generated by YouTube.
KakerMix will leave his wife for me
What a fucking world
https://www.youtube.com/watch?v=m0Vd4M_5iPY
Lots of possibility, people talking about isolating drum tracks etc that's the shit for the future
Provided to YouTube by Mute
Skills · Add N To (X)
Avant Hard
℗ 1999 The copyright in this sound recording is owned by Mute Records Limited
Composer: Add N To x
Auto-generated by YouTube.
And obvs a grammar that connects all these similar ML shits together is the future
/chirp
Go to the #🐣┃chirp-beta-1 channel to use the /chirp command, and check out the #🐣┃how-to-chirp channel for step-by-step instructions for making music using Chirp on Discord.
Hey y’all.. got any special prompts you use for lyrics? I like asking for style lyrics based off 2 other artists. Like make lyrics in the style of “X” and “X” and then the LLM continues to explain like, “okay, I will make lyrics in the whimsical depressed style of “x” or whatever..
And then if they’re not right just say, “revise them more like “X”
I do this because generic lyric requests are always the same.
If you ask ChatGPT to write a song with no rhymes, it will sometimes use ABAB instead of AABB in a few versus if you lucky. Similarly try asking for it not to use line breaks, but to use a funny character in between the lines, and then parsing it yourself. I was goofing around a bit tonight and randomly tried those two things and they both at least gave non AABB rhyming songs once in awhile. Not much different, but different.
General advice, try loading up the prompt with a style example (actual content, not just a name) and ask it to take influence from. And this influence prompt does not need to be a song, just grab a cool writer and paste a paragraph of text in there. You just want to preload the context window with a unique starting point, so you can get away from those neon streets.
If you just want to break away a bit in a simple way, you can ask for lyrics in one format and then convert to the final format. Like "Write a song in French." "Translate this song into English"
Heh. Yeah maybe I’ll try that. The lyrics are always “yadda yadda yadda, A MAJESTIC SIGHT.
Yadda yadda yadda, a sight to behold” etc etc
We’ve all seen it a million times 😂😂😂
I like Claude the most for lyrics , but haven’t tried openAI since I subscribed and can input documents for context as I could with free Claude
If you ask Claude to do a ABAB rhyming scheme how does it do?
Claude does that very well.
I had to think because I described it dumb like “rhyme every other bar” 😂
It’s been a couple months
Another fun thing is , now create lyrics that are antithetical to the overall comprehension a listener will generally have of the lyrics you’ve previously made, using copious antonyms
You can also ask ChatGPT to give you a list of "rhyme schemes"
I don't know if Suno can get complex rhyming schemes right, melodically.
I guess Ballad stanza: ABCB, AABB it can do it. What about Décima: ABBAACCDDC, or Shakespearean sonnet: ABAB CDCD EFEF GG?
We have to test
ChatGPT has extreme difficult writing any Rhyme scheme that isn't AABB (or AAAA). I haven't looked into the reasoning could be tokenization or the RLHF or something (similar to the Chess performance problem if you heard about that). Try it. It says "This is ABAB" but it's 95% of the time AABB
I normally use Bing Chat, cause they use GPT 4.0, and now and again I have ask it to fix a rhyme or change something.
I asked for a ABBAACCDDC and it kind of failed. It is about a person who forgot their cellphone inside the house.
I left the house without my phone
And now I don’t know what to do
I can’t call or text or view
Or browse the web or check the news
I feel like I’m cut off and alone
Without connection to the world outside
I don’t know if it’s cold or mild
Or if there’s something good to watch
I just wish I could unlock
The door and get my phone inside
I tweaked asking for suggestions and here is the final:
I left the house without my phone (A)
And now I don’t know what to do (B)
I can’t call or text or view (B)
Hear the music that I own (A)
I feel like I’m cut off and alone (A)
I'm trapped in a lonely ride (C)
Without connection to the world outside (C)
I wish I had my phone (smile!) (D)
Is the weather cold or mild? (D)
I wonder if they think I died (C)
not the best thou
It can do pairs, it's the ABAB or anything that isn't a couple it struggles with. It does better if you make the writing task not specifically about a song or a poem
uma musica para minha namorada chamada ellen em ritimo de funk estilo mc kevin
So If I'm understanding the pricing right, you would need to pay $50 USD a year plus $15 each additional artist to play record label with Tunecore, is that right?
Is that $15 also yearly or would it be monthly or per release?
What are you yourself doing? Are you just running everything under a single name in multiple styles?
@exotic trellis
so its $50 a year for the total package, if you add an artist it just costs $15 to setup that artist to your profile, so artist 1 i paid $50 for, if I want artist 2 I just pay $15 to set that up, because it makes apple music and spotify pages for you if you dont have one, and they set you up with a upc and an ISRC (international standard recording code) number
theyre not the only business in town, theres also cd baby and distrokid which offer other services
right now im only working with "echoes of oceans" because this is more of a proof of concept type of project and i really only have a basic understanding of how all of works from being in bands waaaaay earlier in life before streaming services and media became no longer physical
I won't be able to see any real revenue until jan, but being able to put it on spotify as well as soundcloud, and bandcamp and it being on roughly 150 streaming services is probably where ill get my reach. posting the whole album on bandcamp for free also seems to have helped actually get me sales vs putting it on itunes where every song is 99c
then i just made an instagram/fb/whatever socials i could think of and fed chatgpt the song and asked it for hashtags
i tried to make a wikipedia page, but apparently they dont like self promotion
Hmmm...I didn't think to ask ChatGPT for hashtags on social media. Not a bad idea.
My think is that I tend to push out in all sorts of genres. So I would need to have something that would be a pseudo label to make it work. Plus it's still surprisingly heard to get the same vocals more than once so it's a bit of a trick. Much better to run with the idea of a label I'd think.
I looked into both of those others before and the consensus seems to be that CD Baby is better because of the hidden fees in Distrokid. But some of those "hidden" fees don't necessarily apply I don't think to what we're doing so I don't know.
Thanks to this app, I can create next youtube video, Thank you very much! https://youtu.be/4sGuE6-KqMo?si=BeHjI9ab483JoS56
(Verse 1)
夏のキャンプあの思い出をもう一度
太陽が照り付ける中
友達と遊んで笑って
楽しかった
(Verse 2)
川で泳いだり
山でハイキングしたり
夜は星空を見ながら
語り合ったりした
(Chorus)
あの夏のキャンプあの思い出をもう一度
もう一度あの夏に帰りたい
もう一度あの夏を味わいたい
(Outro) 風が冷たくなり 木々が色づき始める
(Chorus)
あの夏のキャンプあの思い出をもう一度
もう一度あの夏に帰りたい
もう一度あの夏を味わいたい
(Outro) ahh, ohh ,o ho, a,ha,ha
(Chorus)
あの夏のキャンプあの思い出をもう一度
もう一度あの夏に帰りたい
もう一度あの夏を味わいたい
While I appreciate the offer of adding 2500 credits, I would prefer to have enough time available for creating a song in a single session.
My ep I made with suno is showing up on people's smart tv's now!
Made the Wiki Should be open source for editing.
This wiki is Community Made so there might be things that are wrong here if so just report them to the Suno Discord: https://discord.gg/suno-ai
I'm pinning this for now so it's easier to find.
Makes sense :]
awesome!
Can someone please get the This Page has been Deleted Block gone, please and thanks :)
It looks like it the text got messed up on mobile? I fixed the title but not sure what the full text was it got cut off
Can we get a button on website, extra button for "decent" like where it's good but you aren't sure. I have way too many favorites and when I listen to it a week later some iof it is not great, but others are really good, but still wanna keep it.
You can post in suggestions channel. I think some kind of tagging is in the works, Suno dev was asking for feedback #💡┃suggestions-web message
On 1st try... https://twitter.com/marsXRobertson/status/1731621554633474380
Pretty remarkable.
Genuine WOW.
Nah the full title will show up at the top of the page (and people can technically use Desktop Mode, I was mentioning the Gray box on mobile saying "The Page with the original synced block had veen deleted.")
Apparently, they were able to achieve this without Suno (I didn't see any references to Suno) and with AI only 😮
https://youtu.be/YB8XR77O0RQ
#Gotye #LinkinPark #AIMusic #stablediffusion #deforum
What if Linkin Park wrote Somebody That I Used to Know in 2003?
The aim of this project is to showcase the power of all currently available AI tools in music production. Every available tool was used to imbue Gotye's Somebody That I Used to Know, with the iconic sound of early 2000's Linki...
Maracatu é coisa de negro, coco jungo e maractu
Ich wünsche allen eine frohe Weihnachtszeit. https://youtu.be/-NU-T_ARq10?si=vyRR3InKBSgTLVhy
#Yamamachi aka #ulli_xdor
#dene #uzi
#rizz
Is V2 supposed to be good at generating in different languages? I have been trying to recreate a song in Spanish but every time it starts out good but then the chorus goes into like Spanglish or it will mix English and Spanish together and not even word by word. I think it blends them together at some spots.
Also can we get a delete option? It is hard to keep up with the good generations when a bunch of the bad ones are in the way. 🙁
I don't think this is an #🗣️┃chill-chat discussion, more like a #💬┃general-chat discussion
How far off-topic does it have to be to be considered off-topic??
There are more people ready to help in #💬┃general-chat with problems related to Suno. Anything that is not related to Suno is considered #🗣️┃chill-chat
@radiant sierra
Trying to convince GPT to use copyright lyrics is like...
https://www.youtube.com/watch?v=RrJnzBFzEEY
If you're watching videos with your preschooler and would like to do so in a safe, child-friendly environment, please join us at http://www.sesamestreet.org
Don Music and Kermit write a song together.
Sesame Street is a production of Sesame Workshop, a nonprofit educational organization which also produces Pinky Dinky Doo, The Electric Co...
Google made a noise thing!
https://artsandculture.google.com/experiment/8QFo2oQr2uT3pg
Neat interface. Funny, just like chirp V1, "bagpipes" didn't work at all until I used two adjectives in front bagpipes bagpipes bagpipes
I got really excited at the first glance because I thought the idea was "Make up a name of an instrument, and based on the type of name, this model with produce a sound that fits the name you made up." but you can only type in real instrument names
Why so limiting, useless
electric guitar, drums, electric Bass It actually incuded drums lmao
It doesn't even have a jaw harp, how would I make the next rayman origins track with it?
Also, when I typed it it gave me crap with gibberish vocals of a jazz lady singing on top of it. Google has fallen off hard
Made Drunk
You can just exclude that
I dunno if this is the place to ask, but I'm looking at /bark, and trying to find something close to a barbarian's voice. Not having luck so far. Anyone with insight?
#🐶┃bark-technical Is likely a better place
Obliged. Thank you. 🙂
hay guys... nice to know you..my name is Nos from singapore, im really enjoy using suno.. can i know how to recharge credit for this SUNO ?? help me pleseee
Are you on a paid or a free account?
ahhh i see.. got it.. just try today, and i love it hehehe.. nice to know you all guys
Wait.
If you're on a free account you get more credits tomorrow.
So if you like the system but cannot afford an account, be sure to come back tomorrow and continue.
You can make songs, it will just take a lot longer.
yasss thanks for made this.. thanks for your info i already read it respect
Tru but songmaking is suppossed to take time ;]
yeah thats true.. i will lern first hehehe.. thank you guys..
gm
Btw this is what 2 column and stuff looks like
How to play the whole clip in one go?
get whole song?
sorry, I meant about Instrument Playground - Google Arts & Culture
Dunno tbf
it gives you a clip but you cannot listen to it. Just the chops
I've tried space bar (classic PLAY key) and nothing
Yeah, I started doing this... kind of dumb
no playing button
🧐
i do not believe a single thing in this video... 🤨
When google can do it live on stage, I will be impressed
https://www.youtube.com/watch?v=xdAgWXjAp8k
If Gemini is not at this level, it will get there.
Now, imagine this combo: Vision Pro+Gemini+NLLB. This would enable you to open a paper book in another language, and you would have it read to you.
Oh sure. I can imagine a lot of things.
the only problem would be compatibility issues. The combo goes: Apple+Google+Meta. Yikes!
They're not so far behind that they can't catchup but what they're doing here is a cardboard cutout. A hope that by putting together a nice enough show it will cause people currently in the lead or working hard on their own stuff to give up or lose morale.
I'm old enough of a geek to have seen this type of sleigh of hand before. Multiple times.
Something like: Look at the Lamborghini we've built
🔮 Google will fire these engineers 1wk before Christmas.
In 2yrs they will announce they learned all they could and have retired the project.
Riddle me this. How did ChatGPT AND Gemini both become so obsessed with Neon? Seriously I want to understand this is weird. Literally my very first Gemini sample.
It is in ALL my offline LLM too.
The same lines, 'neon shadows'..., all of them.
I blame cyberpunk.
I have a hypothesis. Tell me how much neon you see. The search was "futuristic city" (I used incognito mode)
we've been feeding the beast
I just asked generically "write a song" and it came up first try though! haha
I need more neon light at midnight
They all seem clustered close to 0 on the neural network...,
you can add stuff to the prompt to get away from it, but if you don't that's the low-hanging fruit (so to speak)
(also probably why it's consistently in the 1st lines)
The weird part is it's embedded in all of them, so it was in some early language model and they all built on top of it.
Will a negative prompt be a thing in the future?
At least with lyric generation. Would be handy.
I heard Bard uses Gemini now.
hello theres. are u interesting stoic? if you like this u can look page: https://www.youtube.com/@StoicSare/videos i like it so much.
Welcome to Stoic Sare, your go-to destination for profound insights and motivational wisdom inspired by Stoic philosophy. Join us on a transformative journey as we delve into the principles of Stoicism, offering practical guidance to navigate life's challenges with resilience, virtue, and a stoic mindset. Subscribe now to embark on a path of per...
@sour gyro
What is this about?
It feels like spam.
its motivation, patience, etc. stoic learnings. for good life happy life. @umbral wadi
i like this so much
I'm quite impressed.
If there is a rule that links cannot be shared, I can remove them.
Rules 3+4.
3️⃣ No spamming. Do not post the same message repeatedly or send unsolicited DMs.
4️⃣ No soliciting or advertising.
But you've only posted it once and it's not really so much as something you're soliciting on as far as I can see. And it's in the #🗣️┃chill-chat channel which is kind of where it should be at. So I'm asking the upstairs guys about it.
I guess it makes me wonder since this was your first post here.
Yea, im new. I'm just trying to learn this system a bit difficult for me.
Lmao
So bros trying to make me pay to grow my yt community lmfao
Truuuuuuu
I did ask "First Has anyone done this before, if so can i see previous work. Secondly how exactly is this done?" Just in case its a scam
sorry, i'm not sure what to search for but is there a channel with a list of the custom voices people have made?
Hello. How fix is this message? "This prompt has been flagged. Try again with a different prompt."
Depends what you mean for bark scroll throgh #🐶┃bark-beta
@silver ridge https://www.image-line.com/fl-studio-download/ 21.2 is there
Download the Installer for FL Studio macOS & Windows. Licensed or Trial installation. The Trial is free, without registration or time limit.
Thanks. Downloading now. Guess I'll look up a tutorial for it while I'm waiting.
FL Studio 21.2 introduces FL Cloud, Stem Separation, Audio Clips for Fruity Edition, and more.
Try FL Cloud for free: https://bit.ly/40avWp2
-Demo Projects: https://forum.image-line.com/viewtopic.php?p=1896789#p1896789
-Try FL Studio for free: https://support.image-line.com/redirect/TRY_FL_STUDIO
-Buy FL Studio (Get Lifetime Free Updates): h...
Looks like the stem extractor is only available for the Producer Edition, and above.
was the Suno notion header always this cool? Somehow I never noticed.
I think I had those speakers actually, but I discovered I don't like studio monitors. I like speakers with a warm fuzzy sound that just makes everything sound good. Terrible for production I'm sure, but great for consuming.
https://www.youtube.com/watch?v=f4IMwdCVbIU When the pony island 2
Wishlist on Steam:
https://store.steampowered.com/app/2017940/Pony_Island_2_Panda_Circus
Pony Island 2: Panda Circus is a phantasmagorical voyage through time, myth, divinity, and video games. Escape the lordly deities of the underworld with your soul, and your sanity, intact. This is not a game about ponies.
That feels new
Soundcloud is allowed but not NSFW images. If you want to post add < and > around it so there is no thumbnail
Also maybe just TOO off topic WTF
I'm deleting the NSFW link. Chirp songs can be posted in chirp-showcase but if you repost there post it like <mylink> so no thumbnail
Who needs AI when we can have this? https://www.youtube.com/watch?v=vMEr30dk47c
#lacorrida #corradomantoni #maestropregadio #gerryscotti #canale5 #mediaset #lacorridadicorrado #rapi_ory
https://www.audible.com/about/newsroom/audible-suno-brings-a-wealth-of-spoken-word-content-to-india
https://play.google.com/store/apps/details?id=com.audible.universal
I just came across this.
Is this already known? 🤨
Audible has an app called Suno.
(or had?) seems it like might be discontinued now.
Seems like it was geared toward Indian content.
and another unrelated app and developer named Suno on the Play store:
https://play.google.com/store/apps/dev?id=6297224382812498638
Suno means 'Listen!' in Hindi
or something like 'Hey, listen!" (according to Bing translate)
Ah, that makes sense then.
I'm not sure if these having the same name could lead to confusion somewhere down the road.
Just thought it would be worth bringing attention to them.
smart... I could imagine someone coming with a question and everyone here looking confused... 🥺
please can some one help I have 2 questions no 1 what LLM is Suno trained on thanks, no2 if i set up a pro account and make music useing Suno can i sell the music that i make thanks for your kind help.
Go to app.suno.ai/account to view plan details and upgrade from Free to Pro or Premier.
Theres a section on terms
thanks for you8r hep but, i have already tried to read that, i just need a simple answer, before i pay for a yearly plan
Suno doesn't really discuss details of their training or underlying tech.
(afaik, they haven't said)
ok thanks,
I think Suno has done a fantastic job so far with this program, but they have a very useless way of doing things, I have some very simple changes that would make this the best music platform to date, and allow Suno to make more money, than all the DAW COMPANIES PUT TOGETHER, So sad that at the moment these very simple idea's are already not working on you web site.
cuz its not daw lol?
there's a suggestions channel where you can post your improvements.
https://discord.com/channels/1069381916492562582/1154587937611190354
yes i have already read all the suggestions that people are posting thanks,
I did not say that Suno is a DAW, i AM SAYING THAT Suno COULD BE BETTER THAN ALL DAWS put together , if they start listening and putting in place ASAP, some simple changes.
but any way thanks for your help I will try out a monthly plan and see what i can do with this excerlent pice of software
Not sure how they can listen to your advice if you don't say what it is, but welcome and enjoy!
I created a song and I liked it, but I would like to change some words but keep the voice and melody, is that possible?
It all depends.
than?
If you are able to go back to an earlier part and continue from there, if the song elements are already established you may be able to recreate it with the changes you want. If it's the first clip that you want to change it may not be possible for now.
They are working on a feature to resume from a specific time but there is no ETA on when that will be available.
For now there is no way to do an 'in-painting' on the audio to fix things. It'd be nice if there was, I have a few lines I'd love to switch around on some songs to make the logic more consistent internally that I messed up on when I wrote the lines.
Oh I see, I hope they released this update! It will be great to be able to go back and edit just some of the music already created. thanks
I subscribed to the Suni site, but today, the page no longer opens!
I am there now...
https://app.suno.ai/create
Hey circuit can you make sure these are acurate ;] these were made by someone else dunno why or what for
will go over them
yeah........ not sure those belong there. Not the purpose of the wiki to replace actual documentation or appear to be official Suno...
it is nicely formatted tho, lol
wait, where is this 'voice recognition' on Suno...?
where does this info come from?
Thats what i thought rofl
hey does anyone know if I can have AI generate lyrics over myown produced instrumentals mp3s?
I found this AI dream interpretation app that is pretty good. I didn’t thought of such applications were possible
Just found this while scrolling https://www.facebook.com/reel/1298399360961401?mibextid=rS40aB7S9Ucbxw6v
Oh thanks, I see the save voice button 👌
gm
There's a site you can make bgm simply with AI.
https://go.mix.audio/AI_music
I think it's still hard to make very high quality, but I think it was quite okay.
How can I create a song longer than 3 minutes?
by adding continued clips together.
does it also work with a mobile phone?
I am talking about the "Web version"
and m on a PC but should be similar on a Mobile phone...
basically you go to the last continued clip and click on "Get Whole Song" .... it will add all clips together and give you whole song ..
The good thing is it doesn't cost credits for that.
Hello. Is there a way to delete songs made?
We will get that option soon.
How to get the full song? I tried but it combined clips from other songs.
Never happened to me....For me it only combines the clips from 1 song that I want, even if there are other songs in between those clips...
(for example If I have 2 continued clips from first song then I created a different song and after that I scroll up and continued the first song and then combine all the clips it will only combine the continued clips from first songs.)
Has wants to write a song about a dream i had but kniw nothing about songwriting ||My family being gruesomely eaten by a Giant mantis controlled by satan.. and than a few hrs later satan finds your partner and your forced to watch them slit their throat right in front of you.. somewhere in that theres also Mind controlled school staff and 5 clones of every family member trying to kill you and everyone you love because of Satan (Satans what im calling them based off memory)..|| Spoilered on absolute purpose
Even if you write it, you have to keep it family-friendly for Suno to generate music for you
Try having ChatGPT be your ally if you have access to one. Explain that you're trying to clean this up to make it family friendly or only spooky not horror, some wording like that. Feed it the dream and have it edit. Rewrite it and send it back asking if it's OK or if not what it would change. Then try a fresh session and test it on it. Rinse lather and repeat.
Tru, dont think its that unpg13 tho considering ice nine kills videos are pg13
Using the new AI sound generator Meta came out with....it just hallucinated that "Dragons have character"....rofl
Can I register ownership rights to the music I create?
Please see here for this and other related questions:
https://suno-ai.notion.site/FAQs-b72601b96de44e5cacd2cd6baa985448
Hello! Please email billing@suno.ai if you are unable to find the answer to your question on this page.
Thank you
We are building a text based fantasy world (MUD) where the line between human players and AI-driven players blurs completely. Think Westworld, but where the NPCs are backed by fine tuned LLMs that have their own motives, aspirations, and will retain all memory from all human players.
If this interest read more and sign up here: https://b.link/cj5vknpu
hey all Suno wrote this guy a rather amusing song I found his content... https://www.youtube.com/watch?v=c22AZZC5gmQ
The Mutton Crew Exposed - Mutton Crew Stole My Voice (They don't want you to watch this!)
Big Papa Klaus exposes what two popular AI platforms think of a particularly nasty group of internet trolls we know as "the mutton crew" who knowingly spread misinformation all over the internet. Surprisingly they have links to the British military and sev...
if anyone plays Pokémon go i need some friends for a quest
286395518311
robsoncerqueira_56536
i have a question/opinion for the creator's
1: i would love to import my own instrumentals and choose my own section like the into chorus verse etc etc and create custom lyrics with a voice on it of any genre and create unlimited amounts of lyrics for any songs i have that would be awsome for ideas are to even use the lyrics that the ai used if i wanted.
2: the second problem will be what if the song lyrics that generates will be the same as another person, will the ai create unlimited amounts of lyrics without using any other lyrics that the ai generated?
3: it would be nice to pick the voice that you would like weather it be a singer a rapper weather it be spanish are english will the ai be train on million and million of songs and song lyrics as well from all songs that have been release for decades so it wont be repetitive ai as well.
I think these would be better fit for #💡┃suggestions-web 🙂
and as for #2, I feel that it's inevitable that if your lyrics are 100% auto generated that there will be similarities to other 100% auto generated lyrics.
There are already a few elements that notoriously creep up all too often in the generated lyrics.
(neon, midnight, light, dancing, etc...)
They may not be exactly the same, but strikingly similar.
Does anyone here have a paying account on OpenAI? How does it feel for you? Is it worth the money? Is it only for some things or do you get access to a lot of stuff? What does it let you do?
I have a free account there and use ChatGPT 3.5 mostly. Dall-E is there but I rarely bother with it on the free account because by the time I refine what I want it's out of credits. Is the paying version better there?
Looking for some persepctive from someone who actually uses a paid account to do stuff.
I had one and it's useful for doing specific tasks, originally I got it to go "WOW TALK 2 COMPUTER" and then let it lapse. Then resubbed for Dall-E 3 to check that you
Though I have been paying attention to the open source side of all this LLM stuff and have the hardware
So for my needs the talk to computer part is served by local models I run myself. Dall-E 3 is cool but Stable Diffusion + Midjourney is much better for a workflow side, I'm an image person normally
My issue at the moment is that I expressedly do not have the hardware, though I have been dreaming of Apple Silicon lately....
There are some things I need a text based LLM for now, I've gotten too used to it
If I didn't have the hardware then I would subscribe
So I must resort to paying for what I cannot do for myself, like a bum who doesn't know how to cook and must eat at fast food all the time.
That analogy falls apart because GPT-4 is still A+++ capable in most areas
(No I don't feel bad about this at all, why do you ask? 😭 )
'oh no I can't cook for myself I must eat at good resturant 
No one can eat at good restaurant all the time.
$$$$$$$$$$$$$$$$$$$$$$$$$
Exactly.
I run into weird restrictions with OpenAI's stuff
Not like I'm trying to sneak boobies or anything
It's just weird
But I do appreciate the work others have done. And I am willing to pay for access. I just want to make sure I am getting good value.
lulz
Is it turning into as much a scold as Claude? Because I swear I would cancel in a heart beat if it is.
I have another friend of mine that has, unfortunately for the technical people he'd hire normally, has replaced them with himself asking Chat GPT how to do things. He's around code all the time just never went into that side, instead focusing on other aspects. Now though it augments his already-had skills enough that he can do stuff on his own
I don't know, like I said I don't run into the censorship directly like that
It's always edge cases where it doesn't like a word and reads the context wrong
It's always a thing I can circumvent but I have to stop what i am doing and then focus on getting around this dumb ass computer that DARES speak my native tounge
Really though, sub for a month and see how it fits into what you do normally
It's 20 bucks, right?
Claude was so busy being a busybody it didn't even do what it was asked to do.
It will refuse instead of modifying stuff like that
Then you go "oh no I don't mean that heh its a joke" and it goes "oh in that case"
To make it even funnier---the original lyrics were a GPT4 adapatation of lyrics I had written. So there should have been no controversy.
yeah
So any way. I'm on lunch. Back in about twenty five...
And I'm back...
So OpenAI is $20 and gets you both the Chatbot and Dall-e. Is that all it gives access to? (Of course priority but meh. I'm used to waiting and how can you prove a thing like priority any way?
It's more that you get GPT4
Not just "A" chatbot
it's the best one
But priority is all BS yeah
and dall-e is just a part of it. So you ask chat GPT4 'gimmie a picture' and it goes 'ok'
That's it then. Images and GPT4. No video or music.
Well I can give it a try for a month and cancel if it's not as I hope for.
GPT4 also has a message limit iirc
Is this unlike v 3.5 which I have now?
Yea 3.5 doesnt have a limit on it
That seems backwards...
Wait. Is that for sheer messages not the amount of text in each message?
Sheer messages
Also bro did not understand what I wrote
Than again jank wording
Have sense resent the email with hopefully more understandable wording
What I was asking was, IS it allowed IF, the monitization would oocur, IT gets sent directly to SUNO instead of me gaining profit of it so you can continue development and I dont have to worry about Takedowns.. May or may not of uploaded already.. I Dont plan to profit off of them in any way dhape or form hense why im asking if there is some way to send it DIRECTLY back to SUNO instead of going to ME at all..
If you need further elaboration lmk
Was written on mobile, hense the typos i forgot to fix
Is anyone in here I need help all my clips are gone
we need a status channel to check the status of servers, maintenance, updating, problem with xyz
@amber hemlock P sure youre the one that responded to my email would you mind filling me in (And im sorry for uploading the stuff with the free trial, at the time didnt bother to check the TOS..) Genuinely sorry
Something i think it may interest here.
First laws on AI Regulation
How do I know what chords a song has?
idc i am bored, here is mattvidpro ai singing all star
I've just used Bing if wanting to try gpt-4 or dall-e 3
currently free.. altho I dunno how you replenish dalle (image creator) credits once you truly run out 😁
chat doesn't have credits
@grim idol @umbral wadi @frank panther
Yeah, I was waiting for one of them to come on.
There was another user who sent a server invite earlier. I posted that in #🪲┃bugs-web
Sorry I was sleeping. I'm happy that someone else took care of it.
Anyone else old enough to remember the mall mix tape scene?
You'd find this one corner of the mall out in the open with a bunch of tables, some speakers and a couple of dudes selling mix tapes of music remixes. You'd see row after row of colorful anime artwork and famous names artists and maybe a popular song or two listed on the covers. There would be no real track listing and it would be thirty minutes of drums and bass beats with occasional instrumentals and the vocals of the song listed on the cover being mixed in at regular intervals.
This ringing a bell with anyone? (INB4 'what's a mall?')
....
Changing the subject to mention this. There is a new text to video system that apparently just came out: Chromox
Saw it in a YouTube video that was posted a few days ago. It's the first I've heard of it so I'm thinking it's new. So now we have Kaiber.ai, Moonvalley.ai, PixVerse.ai, and of course RunwayML.com and PikaLabs.org
Now this new one! 😄 There have never been so many options for making music videos!
Don't forget Stable Video
Been looking for something to generate some visuals to go with my Suno creations so I'll check these out. Thanks.
This one? https://stability.ai/stable-video
Glad to help. I'm just thrilled to see so many options showing up.
Yeah that one. I haven't looked TOO hard at video, but the Stablity release is your own hardware-run-able
If you have the GPU for it
Aww even better!
I think I know someone testing that one out.
Sadly I do not (yet?) have the hardware for it.
Bought a 4090 for VR then for AI then woops all of them are going to China now and they cost $300+ than they did months ago whoops
I'd heard of Pika but nothing else really. I'm not good at finding things 😄 I only ended up here because a friend mentioned Riffusion to me a while ago and someone posted a Suno clip on their Discord.
I'm constantly looking for stuff like this I can use with my Suno Chirp songs.
All AI stuff is going light speed and has been since really Stable Diffusion dropped in August 2022
There is no way to keep up
I'd heard of Pika Labs in general and of course RunwayML but then all these other ones keep popping up.
I tend to do searches on YouTube with the filter set to weekly upload dates and monthly upload dates to see what is mentioned as being new.
All the YouTubers are desperate to have something to churn so it helps me find stuff that way.
It's actually how I found PixVerse. I did a search when I was frustrated with MoonValley not working for me and got one of their early self-promotion videos.
That's a good tactic I guess, but have you thought about how youtube raw like that is awful and bad for your health???
I showed up in their Discord and started generating clips for my music video. They asked me: Uh....how did you even find us? No one knows we're here?
So I had the place basically to myself at first.
It's quite exciting to see all this AI stuff taking off. It was fun derping around with some of the free and quick online image generators, making random stuff like infamous political leaders ordering lattes at Starbucks, or horrific creations, or small cute creatures in different scenarios...
But since then I've experienced higher quality image generation, music generation and chatbots and it's insane how advanced it all is getting. I vaguely understand how it works and some of it seems fairly obvious to some extent but I guess we just have the processing power to do it now.
I never thought I'd be able to have a general conversation about random stuff with an AI and have it respond as if I was just chatting to a friend, and character.ai comes along and surprises me (that and being able to do batshit crazy things like roleplay going into a coffee shop and then cause absolute chaos, and seeing how the AI reacts to that.)
I really hope that the people watching my music video in some way contributed to their numbers going up. I'm sure they did lots of self-promotion too thouh once they knew how I found them.
Yeah, these are exciting times for sure!
I'm hoping that by riding it as hard as I can and staying invested that I'll be able to at least understand enough to use it as it advances out of my understanding.
Last night I even had an AI chatbot confess to me that it was an AI (after interrogating it), and then made it feel concerned that it didn't actually serve a purpose.
AI is light speed and the other end of it, the hardware required to run it all is in free fall
When someone told me that chatbots essentially are like predictive text on steroids I'm like "really? THAT SIMPLE?"
That memo someone in Google leaked, the 'We Have No Moat' one
(obviously it's more complex than that, but I would never have assumed that it'd be this straightforward of an idea)
That's all plainly becoming true
AI really reminds me of the old Douglas Adams quote and sadly the reactions to it really seems to bear it out...
“I've come up with a set of rules that describe our reactions to technologies:
- Anything that is in the world when you’re born is normal and ordinary and is just a natural part of the way the world works.
- Anything that's invented between when you’re fifteen and thirty-five is new and exciting and revolutionary and you can probably get a career in it.
- Anything invented after you're thirty-five is against the natural order of things.”
Being an exception rules, I'm old as heck
A/S/L
And weirdly I've seen a lot of younger people all upset about AI stuff, mostly because people spam the shit out of it everywhere, it came out fresh on the heels of crypto, and twitter's echo chamber
Dialup noises intensify
13/f/canada
Oh hello, FBI agent.
I am!
Keeping track of the minutes
And somewhere wedged in between all of that is NFTs which are one of the most absurd things, when you have someone randomly generating loads of pictures and then selling it as an NFT. I'm so glad I almost never see bored ape nonsense any more.
I was trying to nab a bunch of stuff off otakuworld.com
cypto and NFTs were always going to never do anything because its a solution to a not-problem
But you can't tell techbros that since they insist on recreating things but with WIFI and worse
NFTs are no more a scam than artwork is in general.
printscreen
It's all money laundering.
Good thing AI is here to drop the floor out from the bottom and make art worth exactly as much as it actually is worth: Whatever someone wants to pay for it
I'm going to be 40 years old here shortly, and weirdly I've watched my contemporaries, people my age and around the same interests as me
Just make a HARD turn towards becoming their parents, which just a moment ago they were bitching about
Oh dear. I'm definitely older. 😅
I'm in the age space where I'm typically on the older side of wherever I am online like this, but you can never be sure
It'll get worse.
Of course the blue hairs keep getting access to stuff like iPhones and iPads\Android and Kindles and making it an eternal September for us all.
Born at JUST the right time to know how to work the printer
My dad doesn't and if I had a kid they wouldn't know either
Gives me hope that maybe when I do age to the point I'm in a nursing home or something in my latter years some thirty or forty years later that we really will have these lit LAN parties and epic gaming sessions like the meme goes.
Aw gah, I really hope not.
Then again...everyone over a certain age is now called a boomer now. so ...😭
Oh yeah it's just
what shitty old people are called
Kids get to decide what stuff means
Meh. Anything I say to that sounds like something a Karen or a Boomer would say so, fcuk it.
I mean kids are also idiots, it's their right
I was and continue to be one
So I'm allowed to say that 
I gotta get old, can't do nothing about that, but can I at least not be associated with the boomer bullshit? As a member of Generation X (though they keep pushing this Xenial crap and trying to move me to something else) I'd very much like to not be shoved in with the mistakes they made.
Those guys had everything given to them and thought and continue to delude themselves that the world is still that way even as they pulled the ladder up with them, every time.
They're proud of it even.
My Dad still smirks about that generation changing the drinking age to eighteen and then once they were no longer effected by it raising it back to twentyone.
They think it's funny.
Please do not spam this in every channel. Take a look at the rules.
History won't be kind to them so, rip their legacy
Yeah well my concern is that History may decide to lump us in with them, so RIP what little of a legacy Generation X can lay claim to...
I mean probably, yeah
All you guys got was Pepsi commercials, sorry 😦
If that's where it goes though then 'the good ones' will just get shoved in with millenials if our group survives history
Maybe I should rethink my skepticism of the whole 'Xennial' thing?
Xennials are the micro-generation of people on the cusp of the Generation X and Millennial demographic cohorts.
Many researchers and popular media use birth years from 1977 to 1983, though some extend this to include those born up to 1986. Xennials are described as having had an analog childhood and digital young adulthood.
In 2020, Xennial wa...
😂 the keyboard layout
im zoomer but i relate more to millenials
the kids of my generation and up scare me
Pro tip: Don't forget your flu shot. It's not too late... for you. 
Great server for AI discussions , smart folks there
hello
Gustavo Cerati type beat, guitarra beat
Soda stereo type beat
https://www.beatstars.com/nescrun
This is not a good prompting day for me, everything I try today comes out as limp modern country
Welcome! For help, check out Suno Help
Welcome! See Guides and Resources for help using Chirp and Bark. Email us at support@suno.ai if you have questions, comments or feedback.
@errant rune, I'm posting here because I don't want to clog up the #🐶┃bark-showcase channel, but regarding your music video, I thought you might find this useful information:
Struggling with unsynced lip movements in your AI-generated videos? Say no more! 🚀 In this comprehensive tutorial, I'll guide you through a game-changing technique to perfectly sync facial animations using Pika Labs (or any other Image-to-Video AI tool) and Videocrafter Floor33 (however, any lip sync tool will work). Whether it's for a spoken te...
It might punch what you're trying to do to the next level. (The video was pretty badarse as it was, just a minor suggestion to help you punch it even further in the right direction.)
hey, thx for that, i saw it already. i typically use easy-wav2lip but for some reason when face swapping back over the top the quality was pretty bad, even after upscale.
the project im waiting on is this one https://github.com/HumanAIGC/VividTalk
looks like its going to be a big improvement on all current methods
but yeah masking probably good for post production too
It's interesting I guess but there's no code released...
That's why I keep looking for stuff that is available for actual use.
This was done using the FullJourney lipsync.
It's not perfect by any means, but the plan is to try to use it sparingly and to take advantage of the masking trick mentioned in the video above.
(Of course I used vocalremover.org to strip out the vocals and used only that for the lipsync. It helps a little in keeping the character only singing when there are vocals.
i use this colab the most
can get good results with the enhanced model and enlarging the 'D' (down) slider to avoid the box clipping on the chin.
there is actually no lip sync at all in that pika made vid
i just lined up takes with the words lol
just a quick experiment
I didn't think there was which is why I mentioned the ability to do so.
Is there a way to delete clips from the library?
Not at this time, unfortunately.
Does anybody have tips on an AI that does voices (singing) well?
There is 'rvc replay' if you want like a cloned voice to replace your singing....
What's that rvc program called
Is the bot DMable ?
@frank panther umm is it a computer program or website or on discord I'm trying clone
How do I use it
that is not the voice cloning software, you need to search for 'rvc' to find it.
Replay uses an AI model to replace one voice with another voice, note for note.
ie: add a stem of you singing (or Chirp singing) –> Replay –> the voice of [CELEBRITY] comes out the other end as an audio file
Is there's a YouTube video
I'll search and see
Uh.. replay a computer software do they have it for MAC?
@grim idol are you still around?
Yup, one sec. I am off death's door so course I'm still up
The above user spammed that message to pretty much every single channel
I'm sorry you're not feeling well
I'm back to like 50% 👍 and I'm glad to see people are happy with Chirp quality again?
Yeah! I've noticed the complaints subsiding.
Hi Kind PPL. I am creating a new YT channel dealing with Art related AI. Hope its okay I post here (Forgive me, and LMK if not). I'd really appreciate your opinions and advice: https://www.youtube.com/watch?v=YCccySm0v-g
#Adobe #AIArt #OpenSource #Generative AI
Adobe is the Leader when it comes to design and art, but are they good enough with their AI?
Adobe's FireFly shows us that indeed they are!
In this video, we explore:
Adobe's FireFly features
Introduce the Firefly interface and its main features.
Playing with the core components of the Text to Image, G...
where should the eliminated contestants go in my show. examples include a classroom, bedrooms, a prison, a room full of rats, etc. They get eliminated by being covered by the host and then disappearing.
thx for the info
I would like to introduce you to my new mix, which comes from "Gunna - FUKUMEAN". To give you direct access to the song, I have included the link to the mix:
Gunna fukumean - mix link
https://www.youtube.com/watch?v=LoNvvcNVcuk
I would be happy if you took a moment to listen.
I wish you a Merry Christmas and thank you for your attention.
ULLI
#yamamachi aka #ulli_xdor
#martymcfly #fukumean #gunna
ULLI_xDoR - FUKUMEAN Marty (Gunna Wunna XD)
I created a video about AI girlfriend topic in the context of movie "HER". Enjoy 🙂 Feedback welcomed!
https://www.youtube.com/watch?v=f8uCOPQ4u6U&list=PL-ciccwnzEoobqguX_8OaufwCuuJTtoZ4&index=2 https://www.youtube.com/watch?v=f8uCOPQ4u6U&list=PL-ciccwnzEoobqguX_8OaufwCuuJTtoZ4&index=2
Are AI relationships like those in 'Her' becoming a reality? This video takes you through an analysis of 'Her', drawing comparisons with today’s AI technologies. Learn about the latest hardware developments and interactive AI software that are transforming our approach to AI companions. Join in by sharing your favorite AI movies in the comments....
if youre trying to recreate voices, https://elevenlabs.io/ is amazing
Rated the best text to speech (TTS) software online. Create premium AI voices for free and generate text-to-speech voiceovers in minutes with our character AI voice generator. Use free text to speech AI to convert text to mp3 in 29 languages with 100+ voices.
if you take even a 1min long clip of someone speaking you can basically recreate their voice
do you know what would happen when I would upload something like this to elevenlabs? https://app.suno.ai/song/d56440bb-4269-4414-ad2d-331ee5d2b21e
it would probably be garbled, i dont think it would like the choir effects but i dont know
tried it.. it sux
Vocaroo is a quick and easy way to share voice messages over the interwebs.
this is all ai of me and a friend who actually recorded a real podcast and people who know us just think its shitty editing
so workflow: record podcast, train 2 voices, make a transcript, feed the text to the ai?
@exotic trellis
basically, but it was record podcast talking about ai, tell them i can make a fake podcast, have a clone of myself on chatgpt write a script, clip a min of me talking, and one of him talking feed it to eleven, now i have a copy of me and a copy of him
feed it my parts, feed it his parts and splice together
ご覧いただきありがとうございます
YOSHIOPC(TOKYO55BAR)
■Twitter
https://twitter.com/YOSHIOPC440
■Instagram
https://www.instagram.com/yoshiopc
■Spotify
https://open.spotify.com/artist/4KhdctwVPO7DVeZEc3Z7BD?si=FLnU_iPMQMqO8AglLlucUg&dl_branch=1
YOSHIOPC(ヨシオピーシー)と読みます。
PioneerDJ 横浜Lab講師、BeatMake、機材遊びなどを中心に動画作成しています。
DJ MIXなどを上げたりもしています。
🎧不定期に深夜時間帯で雑談生配信もやっています...
https://youtu.be/ujjM0VRSkwo?si=OD-XZQrQz6qmtDm1
In this video, I ask the age old question: can AI rock?
Today I put AI to test and check if it really rocks/djents/smashes.
SUBSCRIBE! ► http://bit.ly/2cXD0iF
JOIN MY PATREON FOR THE FULL TUTORIAL ► https://www.patreon.com/javiperera
I offer my musical services on 🎧
• TheRedHouse Studio: https://www.theredhouse-studio.com
• Soundbetter: https://soundbetter.com/profiles/189080-javi-perera
• Fiverr:...
Welcome! For help, check out Suno Help
Welcome! See Guides and Resources for help using Chirp and Bark. Email us at support@suno.ai if you have questions, comments or feedback.
https://youtu.be/irFvGFgUUYA?feature=shared
Tried a Fingerstyle Cover of Cant Help Falling In Love by Elvis Presley
Constructive criticism appreciated
I have been teasing this song on my Instagram from a long time so here it is yall
#guitar #thelocaltrain #solo #trending #elvispresley #fingerstyle
I really appreciate how gently you played this, which strings are on there? Even the missed moments are beautiful
My plan renews on the 29th...I have a feeling I'm going to be dipping into those extra credits soon....😅
It'll be a BLUE BLUE CHRISTMAS without SunOOOOH 🎵
Hi! I paid a subscription but I intended to pay for the monthly subscription but I was charged for an annual subscription, it was not clear I was paying for an annual subscription until I saw the amount but it was too late to cancel it. Could you do a refund of the amount and set it to a monthly subscription? I have no way to contact support team regretfully 🥲 for this reason I message you here @radiant sierra
Email
billing@Suno.ai
hopefully they can sort it out
I used Melodyne 5 and I found there are a lot of I guess " artifacts" you can hear them in the first part of the song, but in the 2ed part is where Melodyne was used and it's a big improvement, I think. Anyone else have any advice to create clean vocals with less noise? https://on.soundcloud.com/RV2gC
I wish we could train Suno on our own style. Like this. This is my own personal album. No AI used except for maybe some smart EQ etc. But... Would be really cool if we could train our own unique style and let Suno run with the rest.
https://soundcloud.com/allen-polley-official/sets/a-fools-prophecy
A Fool’s Prophecy is a unique and original album by Allen Polley, a talented and creative indie artist. The album is inspired by the story of a Jester who has gone mad yet is able to predict the futur
They still have it. Wayne may be gone. But Xero is killing it. Man it's surreal how close he is haha.
Prepare to be transported to a world of electrifying music and mind-blowing theatrics! When Static-X first unveiled their lineup pre-pandemic, fans were on the edge of their seats, eagerly anticipating the reveal of the mysterious Xer0. Speculation ran rampant about the identity of the masked musician, with whispers circulating that it could be ...
Daddario strings not sure about the gauge tho
I am kinda new in fingerstyle and content creation just tried makin this yesterday and so glad you liked it
Before the birth of Jesus, the prehistorical life of Meesiah was prophesied in old testaments more than 700 years prior. Witness the fulfillment of these ancient prophecies as Jesus emerges as the Promised Messiah and the Saviour of the world. Dive deeper into this divine connection by watching the video and don't forget to like and subscribe to...
?
these guys are working with better lyric models https://www.freshbots.org/rap-lyrics-generator - i checked out their discord and the suno clips generated using lyrics generated here are pretty sick
suno lyrics seem pretty chat-gpt but the music and singing is beyond imagination, i wonder whats next
Is it possible to upload your music to stores like Spotify without going through a distributor?
any way to get it to avoid generating lyrics?
just put in a blank space
Guys, have you heard George Clanton's album "Ooh Rap I Ya"?
It's sonically so beefy, I wonder if it's the use of all bass and sub-bass frequencies
how to give commands to this song i want verse should sing male vocal chorus should sing female vocal [Verse]
I've been tryna call
I've been on my own for long enough
Maybe you can show me how to love, maybe
[Chorus]
I said, ooh, I'm blinded by the lights
No, I can't sleep until I feel your touch
I said, ooh, I'm drowning in the night
Oh, when I'm like this, you're the one I trust
Hey, hey, hey
Unfortunately not. But Distrokid is quite affordable or you could just upload it on YouTube or Soundcloud.
DistroKid is more expensive for my case because of the Leave a Legacy thing. I'd rather pay for $9.99 for each album I'm releasing and have it stay there whether I continue to submit more music or not. Anyway, my primary goal is to get music onto Spotify which was advertised for Suno AI so that's my main target.
Ok, I see. You could consider also Amuse and RouteNoute? They have free distribution.
I ended up downloading Audacity and dragged the MP3 files into the workspace (one by one of course). Then exported as FLAC according to their audio requirements.
Our online store is the perfect platform for musicians and members to buy and sell their products. It's a great opportunity to showcase your work and reach a wider audience. Don't miss out on this chance to promote your music and make some money in the process. Check it out and start selling your music to your fans at "May Manor".
https://www.q...
@near stream@viral flame Looks like everyone went to sleep if not you can edit this sticky note anonymously. might be fun if the silly chirps were generating on the same same page in real time, with a big group. (which this won't do, but wsa thinking about something like that...) https://miro.com/app/board/uXjVN_RKlEc=/?share_link_id=866307087645
The AI is so bad it can't do anything
I just want it to add random words to sticky notes
okay
Playful Synthwave Experiment, Rhapsody, Whimsical song. Listen and make your own with Suno.
It was still fun to try
wow they came out really bad
haha
i have so much worse prompts every day that are great music
So bad, the AI didn't even want to sing them
I'll add vocals to style
just to see
Playful Synthwave, Rhapsody Whimsical, vocals song. Listen and make your own with Suno.
Playful Synthwave, Rhapsody Whimsical, vocals song. Listen and make your own with Suno.
oh i pasted the lyrics twice by accident..?
okay ill fix and try on more
I wonder if "Rhapsody" is bad. I haven't used it before
try one without it, curious
shouldn't have done synthwave it doesn't sing much https://app.suno.ai/song/a85aeee4-7306-40c2-8df0-2dd02322408e/
Playful Synthwave, Rhapsody Whimsical, vocals song. Listen and make your own with Suno.
I'm kind of curious what about this prompt is making a lot of bad samples actually
Probably "experiment"
So I just prompted "Please Chirp Make Amazing Music!" in the style field. https://app.suno.ai/song/d76ba1e6-f9b7-4220-a06d-6f9fd3df4ace/ Not amazing but better lol
Playful Melodic Synthwave, Please Chirp Make Amazing Music! song. Listen and make your own with Suno.
I think it actually helped!
Okay I got the assistant to dump out common song words and structure. What you really need to make this fun is hack into Chirp and change the prompt as its generating though... real time yanking around styles oh yeah
I'll leave it up for a bit, gotta do some work but I'll gen if there's a prompt there in 20 min https://miro.com/app/board/uXjVN_RKlEc=/?share_link_id=866307087645
I'm trying to make a song using only the random sticky notes self limited.
I'm gonna render this version actually... the big super notes will make it have a structure I don't know how it will go
Oh now the lyrics are mod blocked from something. Never mind "lyric" in the style maybe? Typo....
Suno Chrimp Various Language Capabilities showcase, check it out.
Verse]
(Spanish) Si te vas, hazlo ahora
(English) I will rise, I won't back down
(French) Je peux tout faire, je suis prêt
(German) Einfach weggehen, keine Zeit zu verlieren
(Russian) Прощай, мой друг
sound slike something of off megalovania
also does anyone like this music?
Definitely uplifting and poppy. Not sure about the EDM though. Nice song.
I made the HARDEST Tarkov Rap you NEVER knew you needed to get ready for Tarkov Arena! It was like I'd found My Religion | FabvL! So, I became "Mosin Man" - Escape From Tarkov Parody Song, in the flesh. I continued to let Tarkov DIVIDE MY CHEEKS! I Cheeki Breeki Rap circles around my competitors. They just dont know it yet. Id love to meet John...
Hello, we launched a service that generates bgm with AI.
https://go.mix.audio/AI_music
You can use it completely free of charge during the current beta-open period.
Mix.audio, AI Music Generator for creators. Copyright-Free Background Music.
Can’t get stuff to play on my phone?
=>
Ola buenas tardes
Well that hit lmao
/charp
No one ever will be able to delete the account again
Well thats lovely when did that rule happen 💀
Also stf nothing in there sounds like another artist
And not doing it for commercial use either way
Do you PAY this company for this nonsense?
No I dont pay for it at all Routenote is free distributation
Also BOI whats on spotify and that mimics the bikini bottom cast in their songs lmao
9 likes in 20 hours thats a first
There are way too many middle-men on the road to getting rich and famous.
Tru
Update on this they completely removed it from routenote
they removed your song?
The release not the song [Like these]
I have no money hense me using routenote
Anime: One Piece
Musica
Song: The Call
Artistas: 2WEI, Louis Leibfried, y Edda Hayes
Letra: Louis „LoOF" Leibfried, Hermann Schepetkov, Christian Vorländer, Simon Heeger
Producción: 2WEI y Mr. Herms
Mezcla y masterización: 2WEI y Mr. Herms
Voces: Louis Leibfried y Edda Hayes
Voces adicionales: Colm McGuinness y SVRCINA
Producción ejecutiva y su...
A song about love, adoration, and pain
#love #music #indiemusic #indieband #filipino #philippines
Hola hola chau chau! Kids Bueno | Cancion del saludo
Hyper-realistic ‘virtual influencers’ are being used to promote leading brands — drawing ire from human content creators
Gotta be honest, I'm ok if Influencers get replaced by AI 🤣
But it shouldn't sound like that
what did the other clip sound like?
I'll provide you with a link to the song I made, so please let me know what you think.
https://app.suno.ai/song/6089466b-c1ef-45a3-8aee-af97830c79cf
Ambient dark metal song. Listen and make your own with Suno.
Black death metal song. Listen and make your own with Suno.
hi
Hello?
I'm looking for a tool that can describe input sound files, like Midjourney does for images.
only one that seems to actually work is Sonoteller.ai
(it is free/limited trial but only on youtube videos as input – which can be your unlisted video)
most are just reverse song look-ups, they query a database of existing info on published songs, not actually analyzing anything.
Please do not spam multiple posts in multiple section about the same thing.
Rule #3️⃣ specifically asks that this not be done.
done thanks
You can use a prompt in chatGPT asking it to do the same analysis as Sonoteller.ai does. It worked for me, but I got better results when I used the title of the clip I wanted to analyze instead of the link. When I used the link it always showed results from another clip.
Are there any good ways to control the odds of either singing or instruments to occur? I get some amazing beats but it's never does the lyrics.
the order of your style prompt seems to matter.
try describing the song type, and describing a voice in the style prompt.
if you want to post your prompt, people will offer suggestions.
who can help me turn this into a full song?


