🗣️┃chill-chat

[west hull]

https://app.suno.ai/song/53a82677-ad13-4164-8ea9-663e176666eb/

Im not sure how to use chirp? or discord or anything. Im really new to suno.

[lavish nacelle]

Has anyone tried it with wave music?

this type of music: https://youtu.be/3XEYC4DDMb4

reminds me of smooth criminal but all notes are c

[short flume]

Made a song for Pug lovers and my Pug friend called Habibibi, my little pug. 🫶

Still practicing and working out how to use it correctly but we all start somewhere.

Thanks to suno we can have fun and express in melody

https://youtu.be/x-CxIr9r5UQ?si=zNP--XCfyAwEpw4o

[low epoch]

Cute dog 😄

[short flume]

Ty 😊

[wicked basin]

Hi everyone. As Always, would really appriciate4 your feedback:
https://youtu.be/uTzav2CN5Dg. Thanks to all who reply 🙂

[next breach]

Sounds neat. Try using the style Epic Rock, this is mine when I used it.

Just because it kinda sounds a bit too cheerful for the lyrics lol

TBH it sounds like an anime opening

[hollow drift]

I wish you all a good start into the new year stay healthy and leave love for your loved ones. I hope you like this track. ULLI_xDoR#yamamachi aka #ulli_xdor
#gunnawunna #gunna #ulliwunnasumma
#berlintrapmusic
ULLI xDoR & ULLI WUNNA SUMMA!.mp3 (FROM GUNNA GUNNA WUNNA)
https://www.youtube.com/watch?v=CIu6fCIZbZQ

[naive urchin]

Do you produce AMVs for others?

[sick star]

I found a good bot to make ai cover in discord by simple clicks

[acoustic badger]

https://www.youtube.com/watch?v=pcFG9Sdss3o&pp=ygUJa2lkc2J1ZW5v

[radiant sierra]

Welcome! For help, check out Suno Help

[void mesa]

Yes, hide the eyes. No one will recognize you if they can't see your eyes. 😆 🎉

[fervent mango]

what does rhe remix button do?
does it change the lyrics or the song genre?

[frank panther]

Remix is a copy/paste of the text prompts, but you need to have Custom mode active or it doesn't do anything...

[fervent mango]

ok, thx for the info

[random stream]

Hello, the music I make is usually very short in duration, less than 1 minute. What is the reason for this? Some people have made music for even 4 minutes.

[long tendon]

So, given that I believe there are plenty of Americans here, is anyone willing to have a chill chat about USA politics right now?

[frank panther]

the webapp makes clips 2x longer than discord, and stereo.

for both the clips can be continued with another clip...

let us know which you are using.

not with a man, no.

[long tendon]

It is very fascinating to me. It is also very fascinating listening to the people of this channell called the meidastouch

It is my belief that Trump will be re-elected

not that i support the guy

In my opinion, the current american politica climate it has been a long time coming

[frank panther]

How old is your country?

[long tendon]

Ouff.. Italy it's well old xD

[frank panther]

How's your political landscape?

[long tendon]

Fucked up. But slowly getting there

Europe it's not immune to the wave of far right sweeping the world

[frank panther]

This is why I don't have political discussions with men. So many pointing fingers, so little responsibility for their own.

[long tendon]

Just to be clear, I don't have intention of "pointing fingers" I am more interested in the how a situation got where it is

My current opinion: Personalities like Trump, the Argentine president, Boris Johnson, etc. capitalised on a general malaise that is permeating everyone

By giving "easy" targets:
Things are bad?
Well it's because of the immigrants!
Things are bad?
It's the deep state!

Pair that with a non-stop neoliberal individualistic rethoric that has been force fed trough the education system, creating the illusion that "the world is one's oyster" or that "You can do anything you want"
Now, generations of people are bitter and angry that well.. They couldn't actually do what they wanted

[inland dust]

is this music server?

[umbral wadi]

Political discussion is almost always a waste of time. None of us can change the things we get upset and want to discuss, most of the time those hollering past each other have rigid inflexible opinions on things that cannot be changed by discussion, so all we do is make each other angry. And for what? These are things we cannot change. So why do we need to make enemies out of people who could be friends by discussing things that cannot be changed?

May as well have a rousing debate over what we should do if gravity were repealed for a day. Or discuss in all seriousness what would need to happen once one of us won the lottery, and get mad that this person or that person wasn't spending the imaginary money in the manner in which we think it should. It's just as pointless.

I don't care about winning debates on stuff like this, I'd rather keep you all as friends rather than being right about things.

[umbral wadi]

It is a place where people who use AI (Suno Chirp) to generate music can come to share what they've created, seek advice on what they're trying to create and in general discuss music and how to make a better song with AI and to a lesser extent, how to be creative with music and other things generated with AI.

For instance some of us also enjoy making music videos with AI to go along with our music.

[hollow nova]

do you use AI to make your music videos?

[umbral wadi]

I do.

https://www.youtube.com/playlist?list=PLOqtzKnPd6XdsQcPSYgLiix_QQ-hPA0oI

I have a third one which will be released later this year. And have been trying to work on more besides.

[hollow nova]

very cool 🙂

I haven't seen pix verse yet. I am checking it out now.

[umbral wadi]

There's also Morph Studio: https://www.morphstudio.com/
Chromox: https://chromox.alkaidvision.com/
Pika Labs: https://pika.art/

[cosmic tartan]

I used Pika! I used it to generate some funky effects with my pictures and videos.

[subtle hearth]

@umbral wadi @frank panther @grim idol

[Dont want to ping 7000 times but theres one in all channels

[opal folio]

howdy all, I fell in love with Suno recently, so bought myself a subscription and set myself the challenge of creating and publishing a track, so I used chatGPT to help write some lyrics, used suno's custom mode to create the base track and then used Logic Pro to edit until I was happy with the track. If you want to check it out the end result is here on spotify https://open.spotify.com/album/4PX90lkH7MWA6RRdJNnHBm?si=XXpsDf4ORHW5awF-dmJ7eQ, a fairly wonky slice of pop that makes me laugh every time I listen to it 😄

Of course that's only half the fun, so I had a tinker with Midjourney and Stable Video Diffusion and made a video to go with it https://www.youtube.com/watch?v=1y6Js0gP9Wg. Nowhere near perfect but makes me happy.

I've been lurking here for a while and not said it before, but none of this stuff would have been even vaguely possible without all the cool inspiration here and all the people who have shared what they know. I'll try and pay it back and put together something that explains how I did the specific parts if anyone is interested.

[north crown]

is there any way that i can use this service by API?

[lyric rover]

oh thats cool

i also didnt realise there was chat, is this new?

[terse spade]

https://open.spotify.com/track/3JoyHpOHnqKBpbIh1sPWLV?si=deb2e18599344067

[umbral wadi]

That was awesome! 😄

[west hull]

https://www.youtube.com/watch?v=9dgC3kP3FAo

this is so awesome! the lyrics are almost nonsensical( i wrote them, lol) but it plays so fucking weLL!!!!! damn im just so impressed..

mixing lots of genres comes out interesting as well

[umbral wadi]

I hear people complain about the AB rhyming scheme of songs generated by AI and then I come across something like this....

https://arabkitsch.com/business-directory/150/after-a-thousand-years/

Twas a thousand years ago, We met, I know,
Where the tropic breezes blow, I loved you so,
Do you still recall the past, When last we met?
It seems like a dream That I can't forget.

CHORUS:
Back where those Egyptians
Used to carve inscriptions,
There, upon the temple wall,
Long ago I found you,
Roses all around you,
Then I heard you call, by the moonlight,
We were Oriental Sweethearts sentimental,
Till we had to part in tears;
But love lives on forever,
Here we are together,
After a thousand years!

Till the Pyramids so old Shall fall and mold,
I will claim your heart of gold, To have and hold.
While a thousand years have flown, My love has grown;
Though fate made me wait, You are still my own!

[frank panther]

That's some Tin Pan Alley gold uh... tin.
LOL

[umbral wadi]

Hahahaha...

Yeah well it's one of the 'lesser known' public domain songs for a reason. But I'm bored and after I get AI to replace a troublesome line I'm going to see if I can spin this straw into gold...

[frank panther]

(needs neon)

[umbral wadi]

For a first run through quick and easy it's not too bad...
https://app.suno.ai/song/bbc2c6a7-3fc8-4099-b7d0-1a06acb07f7a

I will now endeavor to add neon.

[frank panther]

this is super cute 🪩 😎

[umbral wadi]

There's some gold in that there straw.

[stoic notch]

the vocals are not that clear but i like soul/jazz songs

[obtuse ingot]

https://youtube.com/shorts/QON6jLnRrng?si=wE75VZl3zqS_qXmp

[west hull]

https://on.soundcloud.com/2YtMMvRN1oJtcbP47

[deep jasper]

lol peanut butter on pickles 🗿💀

[void mesa]

It's delicious. You should try it. Dill, not the sweet bread-and-butter kind.

[livid drift]

thinkin bout subscribin, anyone know any promo codes?

[umbral wadi]

As of yet there have not been any promo codes for this AI.

[livid drift]

i dont understand, when I had the free version, makin songs only cost 1 credit.. now its costing 5 credits a song

[umbral wadi]

I don't believe songs were ever only one credit.

[livid drift]

wait they were 10 each, now theyre 5

im confused

thats what I meant

whered the 0 go

my bad

my brains being burned out trying to make 1 freakin track be able to be continued... half of them are ending w/ a bunch of silence at the end, another half are eating the first word which is one of the most important words...

[radiant scaffold]

Try to use nonsense nouns, f.ex. vegetable names in custom mode - genre style input. Awesome fun 🙂

[marble kiln]

How do I see every chirp I've created? I can't find some I remember making...

[umbral wadi]

Please do not post the same question in multiple channels.

[marble kiln]

Roger that. 👊

[radiant scaffold]

Library, then go to “next” maaaany times. No last page count. 😉 For future: mark as liked, you’ll not miss it.

[umbral wadi]

🤦‍♂️ Oh Jeeze...did I get distracted and forget to paste the link in??

I'm very sorry. Yes you can find the Library here: https://app.suno.ai/me/

EDIT: No, I see that I responded to your other post here: #📚┃suno-school message

[west hull]

how do i use these AI tools to make the 2024 release of this song amazingly awesome? https://www.tiktok.com/@dainisw.michel/video/7314380894918233376?is_from_webapp=1&sender_device=pc&web_id=7297021507129034273

[west hull]

i have a recording concert booked for Dec 1, 2024 -- the crowd will sing the refrain -- but it would be sooo cooool if AI could contribute to this song as well...

[languid arrow]

(╯°□°)╯︵ ┻━┻

[fierce jolt]

Curious if anyone thought of using suno on here instead of their machine they provide free gpu for 6 hours in the free plan https://www.businesswire.com/news/home/20231213860557/en/Lightning-AI-Introduces-Lightning-AI-Studios-its-Enterprise-Grade-Platform-for-Rapid-prototyping-and-Deploying-AI-Products

[solid meadow]

Replicate.com 👍

[hollow geode]

link?

[west hull]

My mum said

[hollow geode]

about what

[west hull]

Nothing

[nimble temple]

Hi

[long tendon]

Today I was thinking about the state of AI and the trends I see associated with it.
I would love to have more data about it but I personally don't think that an "AI bubble" is unthinkable

[sick star]

https://www.youtube.com/watch?v=OFYP30GwOuQ&ab_channel=Kent
This is the video of how to use it also includes the website.

[pliant forge]

A step to the left of music.
https://www.nfb.ca/film/rip_a_remix_manifesto/
https://www.everythingisaremix.info/

[cosmic tartan]

the play button tricked me

[scenic hound]

In your experience, what's the best voice cloning service currently available? I've been trying ElevenLabs, but I wonder if there are better options? Especially for other languages like Spanish or Japanese

[fresh dawn]

rvc

[scenic hound]

Thank you. I'm going to look into it

[hollow drift]

https://www.youtube.com/watch?v=SEb7oULDvXE

[next breach]

So, the songs I'm making follow a story, and I thought here would be a good place to put it.

Earth is dying due to pollution, and four people step up to lead a space mission of finding a new home. Together, they find Haven IV.

Haven IV was then split into 4 quadrants, labelled as Sectors.

They built a new city that spans the entire planet. The only issue, is that those four people control everything. A dictatorship of sorts.

A rebellion starts to fight back against them.

That's about the sum of it.

[long tendon]

How long before "Open"AI gets absorbed by Microsoft?

I mean, officially absorbed

Like OpenAI no more kinda thing

[leaden halo]

Already @long tendon - microsoft released the CoPilot app - that is just an openai reskin essentially.

[runic forge]

Hello friends, what could be some meta suggestions for a children's song?

[umbral wadi]

Please do not spam posts in multiple sections. Please reread the #📜┃rules and note #3️⃣ which specifically addresses this.

[scenic hound]

Sorry to bother you, but I have a question, if you don't mind. I am trying to clone the VoiceConversionWebUI repository using git, but I see there are several large files. I saw there was an option to skip LFS files so I chose that. Is it still a large download? It seems to take a long time

[fresh dawn]

easiest way is to just go to releases and grab the zip, extract and use that

[scenic hound]

For example, RVC-beta-20230513.7z, right? I see there are a lot of versions

[fresh dawn]

whatever the most recent one is

[fresh dawn]

this for nvidia, this for amd (direct download links)

[west vale]

is it possible to get the tracks in wav format?

[hasty isle]

I don't know if its the best, but it's free an open source, so...
https://research.myshell.ai/open-voice

[radiant scaffold]

No.

[subtle hearth]

This occured recently: https://www.cbsnews.com/news/music-producers-lawmakers-artificial-intelligence/

[valid whale]

yo

[frank panther]

But the human-authored works CAN, hence the crackdown on 'fakes'.

[umbral wadi]

It's also why I have very little interest in doing these AI clones so many people seem to be into.

I don't mind quoting a bit of style or sound, but I'm looking to go somewhere new, not somewhere someone has already been.

The more people seek to put words into someone else's mouth or song, the more of a pushback we're going to see.

[umbral wadi]

....

OK, now this is insanely interesting:

https://www.rosebud.ai/

[viral flame]

It does. I just joined the waitlist.

Good thing I'm patient.

[umbral wadi]

The HackerNews discussion brings up a good point though, how do you debug?

https://news.ycombinator.com/item?id=38868185

lishali88

Launch HN: Rosebud (YC S19) – Turn game descriptions into browser games

Hi everyone! I'm Lisha, the founder of Rosebud AI (https://www.rosebud.ai/). We're building a platform to help users go from description to code to game. We aim to make game creation accessible to non-technical creators, so our UI provides explanations alongside the generated code.Users have created a diverse range of games on Rosebud, including...

Hacker News · Jan 4, 2024, 3:33 PM

[viral flame]

You make the AI do the debugging, of course. 🧠

It created the bugs in the first place.
It's its own responsibility to fix them.

[frank panther]

I'm not sure telling a broken bot to fix itself is going to work out... 😬

[viral flame]

You may be on to something there.
Use a second AI to repair the first one.
Then the two bots an work together on training a third bot for the debugging.

[radiant sierra]

Welcome! For help, check out Suno Help

[viral flame]

Welcoome Jordan!

[radiant scaffold]

WOW

[umbral wadi]

First they ignore you. Then they ridicule you. And then they attack you and want to burn you. And then they build monuments to you.

Nicholas Klein

[west hull]

https://www.youtube.com/watch?v=WL3BGww_bIY

hey check me out

i figured it out

[west hull]

i love this one

https://app.suno.ai/song/408ffb01-0d88-412e-9d25-89a50e6c71c0

im gonna finish it tomorrow cuz i dont have enough credits

[long tendon]

Seems reasonable enough to me

[radiant sierra]

Welcome! For help, check out Suno Help

[sage stump]

https://youtube.com/watch?v=6dtWbeHNRGU&si=DY5FAGARn3tawW2E

[visual scarab]

Not even remotely correct.

[valid lintel]

圣诞快乐歌曲，儿童演唱，大调音阶

/圣诞快乐歌曲，儿童演唱，大调音阶

//圣诞快乐歌曲，儿童演唱，大调音阶

[valid lintel]

/chirp圣诞快乐歌曲，儿童演唱，大调音阶

[radiant sierra]

Welcome! For help, check out Suno Help

[valid lintel]

/chrip

[dreamy dawn]

A grunge alternative tribute to "One Punch Man" using Suno AI.

[strange grove]

https://www.youtube.com/watch?v=hys_MPhtmhI

[lavish plover]

French melodic hardcore / metal hardcore

French folk /baroque song

French hardcore metal djent

[flint adder]

Hows the weather where you are living?

Listen to Chickens on Suno! 🎵

https://app.suno.ai/song/74884733-5c22-41e0-85e2-3333f703e7ff

More chickens to tortue you

[icy lotus]

I'm definitely never going to share that song with my kids, they would love it too much 😂

[unborn bane]

CHICKEN CHICKEN CHICKEN

[flint adder]

I know what you mean! This will make them obsessed!

But a good introduction to electro house music

[subtle hearth]

Guess itll be a bit :/

[frank panther]

what are they asking for exactly...?
Proof that you have a paid acct with Suno?

[subtle hearth]

Likely which presently I dont.. No money and Sunos stated cant pay them directly so yea ill stick to yt for the moment

But it is a shock they just now heard about suno 🤷‍♂️

[frank panther]

it's interesting that they give the actual link to the Suno FAQ page... 👀

[subtle hearth]

For ref I have suno created music on spotify.. I think the new ruling on AI music was the reason for it

[frank panther]

a new ruling...? ... from Spotify?

[subtle hearth]

Nah this

[frank panther]

Oh, that's not new. lol

[subtle hearth]

Ah they made it seem new than again havent been to active about date checking

[frank panther]

It was new publication, new 'guidelines'
(not a new law or decision)

(sorry for pedantry)

but If you've used Routenote before... It must be a new agreement with Suno...?
(I wonder if Suno was involved at all?)

[long tendon]

I would be honestly surprised if streaming outlets were to voluntarily allow AI generated content.
I personally would not expect them to give in on it because of the notorious lobbying coming from major labels

This gives me an Idea:
I suspect that the "Majors" (EMI, Virgin, Universal, Warner, etc.) Don't really have a backup plan for when the copyright of their golden gooses expires

I haven't researched (don't even know if it exists) a breakdown of their revenues

[umbral wadi]

The big media companies always cry 'poor me' any time something new comes along and try to destroy it. Then ten years or so later they act like they invented the new thing and pretend they were always for it. It's a pattern as old as time.

[frank panther]

I'm curious how auto-detection (?) knew the song was from Suno specifically.
🧐

[subtle hearth]

Its manual reviewing and on all of these I credit Suno for composing and Lyricism thingy

[frank panther]

They are saying they 'detected' Suno by reading your metadata that said Suno, and then lecturing you about Suno.

😶

[exotic pelican]

Hey all. Not really A.I. related. But... I do my own original work as well as A.I. fun. The link below is to an album of my original work. It is a mix of Electronic, Circus (ish) like tracks. It is basically my view of the world and based on many dreams I have had through the past 20+ years. Enjoy:

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 future with accuracy. The Jester sees a bleak and grim future for humanity, full of war, chaos, and despair. He tries to warn the world, but no one listens to him. He also tries to summon positivity, but ultimately fails. The album takes you on a dark and grim journey of hopelessness, with songs that mix circus, silly, sad, moody, electronic, downtempo, and trap genres. The album also has a glimmer of hope in the end, with the Jester reminding the world that there is still hope in laughter. A Fool’s Prophecy is a daring and innovative album that will challenge and entertain you.

https://allenpolleyofficial.bandcamp.com/album/a-fools-prophecy

[umbral wadi]

Please contact Suno at [e-mail] about this. They may direct you to another email address. If these files were produced while having a paid account with Suno I don't see why there would be any issues here and I think the team at Suno would definitely like to know more about this situation.

@subtle hearth, just to confirm, are you on a paid plan and were you on such a plan when these songs were created?

[subtle hearth]

No and no as i stated before no money, and I read the TOS on it after I uploaded them to Routenote :/ sorry bout the hassle with legal stuff.......

[umbral wadi]

I see. Then this kind of sucks for you and I'm sorry it happened for you but it's an example of things working as expected, I guess, right?

Kind of answers the question: Who would know. Apparently everyone.

(referring to people who say they can violate TOS and who would know, not you.)

[subtle hearth]

But like genuinely sorry if needed ill find a way to take them off the platforms somehow

[long tendon]

Up until the 2000s I would say that Media majors might have had a "role" in "Bringing new artists" to the masses but nowadays before they even consider you, an artist has to create a following on their own.
I personally find it a bit sad that then people that could have remained owners of their own stuff decide to go with those fiendish corporations (my own opinion)

[umbral wadi]

Which since you were attributing and everything this sucks even more. We really do need a place where everyone (paid accounts and not) can have their music streamed from. I really hope that the "Trending" section gets some love soon.

[subtle hearth]

Yea :] I mainly did the thing for accsesibility so if wanted they can have the songs in their playlists on their platform of preference

Or some sort of non-comercial platform could work mayhaps what I mean is if its made by AI the artist and stuff wont be payed in any way

[umbral wadi]

I did post the link to the Clone Wars github with several open source clones of existing music streaming services. Maybe someone can look through those and find a way?

Link to the link: #💡┃suggestions-web message

[subtle hearth]

Could look into it but rather focus on learning GDscript so I can make the fan game I wanna make

[umbral wadi]

::nods::

Such a thing would take a lot of resources beyond just the front end any way.

[fresh dawn]

did you when you made the songs you published to spotify?

[subtle hearth]

Yea makes sense

[fresh dawn]

oh nevermind

[umbral wadi]

I mean if you went full AI music only and full non-for-profit, there might be a way to daisy chain services like those free online cyberlockers that offer so much bandwidth a month for various accounts and rotate them. I know that some services exist that basically do this now for video. But that sort of thing is way beyond me and I shudder to think of the regular work that would be needed to upkeep it. There's a reason why they all have ads or you risk geting malware at those places. They need to justify the time sink in hard cold cash to keep it running.

[fresh dawn]

could potentially be an audio fingerprint in the music suno generates

[fresh dawn]

as an owner of a nonprofit, I feel like I should clarify that nonprofit doesn't imply noncommercial. a nonprofit distributing music through Spotify, YouTube, etc. would still be doing it for commercial reasons

so id say stuff generated from a free account distributed under a nonprofit would still be against tos. tho I'm obviously not a lawyer or affiliated with suno so

[umbral wadi]

Valid.

I meant non-commercial.

But obviously there are a ton of licensing issues.

[cosmic tartan]

MUSE

[lethal agate]

So, does someone here speaks Hebrew, i'm thinking of using already existing lyrics to make a new song with lyrics that were already used on another song

[viral flame]

I don't speak Hebrew, but I'd be more than happy to listen to your song if you felt like sharing it here.

I won't understand a word but I can appreciate the music.

Is it klezmer style by any chance?

[lethal agate]

What i meant is i would use the lyrics to make an AI song here, and probably i would use the lyrics in another project that i'm working on

[lethal agate]

Also, i also dont speak Hebrew

[fresh dawn]

unless you wrote the lyrics or created them via ai, no

[lavish siren]

what happens if I cease my subscription?

[subtle hearth]

Does suno have copyrighted vocals?

[long tendon]

Tbh feels like ot would have been less hassle to simply say you made it and not Suno xD

[subtle hearth]

Lol

[frank panther]

Is this an ad for their royalty free audio samples shop? 😂

Or are they just trying to steal your vocals?

[unborn bane]

Ask them where they think you sampled them from

[umbral wadi]

This!

As always, if the service is free remember that you are the product

[steady citrus]

@umbral wadi basically the creator of SplitticAI admitted to stealing code and programs from other AI to develope and Build SplitticAI, instead of working to create and build his own.

[umbral wadi]

Oh, whoa! Now that is quite a bit worse than just doing covers. Yikes!

[steady citrus]

Yeah. But not counting SplitticAI (Controversy) or Voicify (fun meme voice over covers) what is the second best program capable of producing full songs like Suno? Assuming Suno has any real competition at the moment.

I'm looking forward to one but it's not available yet

And That's Jen-1

[umbral wadi]

See I don't know of one. And I'm not saying that because I'm a mod here, I'm saying that because I'm not using anything else. 😉

[steady citrus]

True. I think Jen-1 might be a sort of competition. But I don't think it'll be able to provide vocals. Just high quality soundtracks.

[south harbor]

Hello, does anyone know a good service provider of music generation (both vocals and instrumentals) with API access as well?

[south harbor]

How about Splash Pro (previously BeatBot)?

[subtle hearth]

Heya matpats retiring https://www.youtube.com/watch?v=8R1_TqU68yo

[exotic meteor]

I’m sad

My history teacher used this bot to make Obama rap

💀

[lavish girder]

that is straight up unexpected

[exotic meteor]

Real

[lavish girder]

dont want to know what it sounds like

gonna go play cave story, see ya.

[exotic meteor]

LMFAO

[opaque temple]

It is indeed a shadowzone, especially if someone have used their own lyrics - and not an automated generated song text.
Now lets take this a step further, and I take the Suno generated song, cut it apart, and introduce a new section of music into it.
Then it will be a new and unique piece of music that one could claim the right for.
The copyright thing got fuzzy when people started to sample the work of others, singing duo with artists that had passed away etc.
This was solved by applying an already existing rule used for classical music, and covers.
A symphony is always atrtibuted to the composer, lets say Bach. But a recording get a label and name of the conductor and orchestra who plays it.
And is seen as an interpretation of the work, and so copyrighted as such. (Try adding that classical piece to a film on youtube and you will see a copyright strike - following that rule.)
So when I chop the Suno generated music up in parts, and rearrange it, the result is indeed my interpretation of it and should be viewed as such according to the definition that apply above.

[fresh dawn]

1. just rearranging the parts doesn't constitute transformative work
2. even heavily altered samples from songs are subject to copywrite infringement
3. classical is in the public domain, and not relevant

[fresh dawn]

nice thoughts tho

regardless, sunos tos is fairly clear on this topic

[tawdry fiber]

I share the same reasoning as you. This is clear from the Berne Convention.

[opaque temple]

Thank you acamposxp.
And I did indeed read up on the Berne convention at one point when we had a piece that were a tad close to something that already existed and was known.

[tawdry fiber]

That's why I suggested in the specific section on Discord that Suno review its rules of use. The way they are (especially the part about having exclusive rights to what was generated by a free account), it doesn't make any sense.

[opaque temple]

AI generation of music, images and film have indeed caused a new set of problems for the interpretation of copyright.
And the policy for the various engines differs, some gives full copyright for the person who generated it - with a reasoning it's their prompt which is the origin. (And it is!)
While other watermark and have various condition apply. And with the TOS Suno belong to this group. And don't get me started on the Models and Lora's used in image generation.
A mirkland for image generation services, as the images contained therein are both copyrighted works by both living painters and digital celebrities, not to mention faces of actors and public figures.

[tawdry fiber]

An off-topic question about breeding out there. I don't know if you're feeling it too. It seems that AI Cover (RVC) is getting more airplay than AI sound generation with unpublished lyrics. It seems that many people don't even want to bother writing something new or generating it. Legislators are going to have a lot of headaches.

[tawdry fiber]

In addition to the author's right, which is associated with charging for use, there is the moral right. Public domain only ends the first. The second is eternal. It's in clause 6 of the Berne Convention, which is also adopted in the USA. Clause 6 bis of the Berne Convention addresses the moral right of the author. It provides for the protection of property rights and moral rights. Moral rights include, essentially, the right of appointment (or paternity of the work) and the right of integrity of the work, in the face of any changes. This means that the author has the right to be recognized as the creator of his work and to oppose any change that may harm the integrity of the work.

[frank panther]

trading on celebrity likeness and familiarity of (other people's) existing works IS the reason for copyright laws in the first place.

Once 'classical' music could be recorded with fidelity, contemporary orchestral music stagnated. We won't hear another Rhapsody in Blue that moves the bar for contemporary music, instead we have John Williams recycling Wagner and Holst's greatest hits forever.

Most people don't even have a word for contemporary orchestral music, it's 'some kind of new classical' which is an oxymoron...

[tawdry fiber]

Perfect!

[opaque temple]

Since someone mentioned the ID of trax above.
I took a break and had a leetle fun feeding a cover of mine into one of those music id services, it id'd the piece not as one but two different ones.
The first one is obviously AI generated, while the second is a quite nice piano piece.
BUT! Neither result is correct! What I uploaded was my cover of a track by Mike Rutherford from the album Smallcreeps day. 😹

It completely failed to identify two covers of game music I've done, more oddly it also failed on a klezmer style song which I inserted since JawKnee mentioned that style here.
Odd, since the original is included in what could be called a radical remix on my part - with no less than 4 layers but the original is there. Lastly I tried a metal version of Holz I made - where I included a riff from a metal version made by others.
And bingo, that one got spotted.

[frank panther]

AI law enforcement is going to be a new hell for everyone.

[fresh dawn]

you mean like the districts already using it?

OH

You mean like copywrite law and stuff nevermind

[opaque temple]

Yes it will be, can only speak for myself.
But if or when I ever make anything of that stuf, I will give credit. (And I actually do have permission for that Rutherford cover.)

[solar holly]

Since becoming a member, it's been like playing spin games with a lot of credit for completing one song. Every time I click continue, all kinds of voices come out and ignore my prompts. It's like he goes where he wants, plays with various tones he likes, and returns to me with a connection song, then creeps to another and returns to give the mood connection of the next song. It's amazing how Suno.ai builds a business strategy model. With a free plan, I can make more than 10 songs in December. Here is an image on my YouTube, where you can see all of this is under the free plan.

[tawdry fiber]

And it changes from country to country. In Brazil, for example, parody is allowed. Deadlines for public domain are also different. The original version of Mickey Mouse won't be in the public domain in Brazil for another 18 years, if I'm not mistaken. It's 70 years since the death of the last co-author.

[dark fjord]

I'm having the same problem, I don't know if I'm using it wrong or not, but looks like that when I try to continue the music, it's no consistent.

[noble root]

Instantly Transfer Face By Using IP-Adapter-FaceID: Full Tutorial & GUI For Windows, RunPod & Kaggle - Face Embedding Caching Mechanism Added As Well > https://youtu.be/rjXsJ24kQQg

[subtle hearth]

Album covers for days won 2 art contests over at fotor

[smoky locust]

Please tell me how to delete the song

😭

[umbral wadi]

There is not yet a way to do this but it is being worked on, see: https://suno-ai.notion.site/b72601b96de44e5cacd2cd6baa985448?pvs=25#a59361b5f07d4a5d84472624982246b8

[safe bloom]

Hello there 🙂 I hope the devs look at this! I have some suggestions that would make Suno.ai so much better: #1. Add control over BPM #2. add control over the type of voice and pitch (Male, Female) #3. Add a way to select the length of generation #4. make the unused credits roll over to next month #5. Improve the use of the "[section]" commands for better structure of the generations
If you agree guys, please copy this so the devs can incorporate it into the service ❤️

[umbral wadi]

OK. Stop.

Go back and re-read the #📜┃rules you are breaking. Particularly #3️⃣ which specifically addresses this behavior.

I am also giving you a ten minute timeout to consider the rules. Thank you for your understanding.

[safe bloom]

Sorry about that.. I just need the devs to read it 🙂

[unborn bane]

we need you to not post that over and over

[safe bloom]

I get it 😉

[unborn bane]

do you though

[safe bloom]

Yeah... Do you think the platform is perfect as is?

[unborn bane]

that's not what we're talking about

at all

so I guess you don't get it

[safe bloom]

I get the part about not posting the same message in multiple rooms. The reason behind doing so is to get the attention of the devs

[unborn bane]

and if everyone did that, this server would be a mess.

[safe bloom]

Cactus, do you use the suno.ai? - and pay for it?

[unborn bane]

how is that relevant?

[safe bloom]

Getting what you pay for is relevant, don't you think?

[unborn bane]

jesus, you still have no idea what the problem is

ignored

[fresh dawn]

no, i definitely see the issue

[umbral wadi]

What was that reddit bit? summoned redditors?

[fresh dawn]

I've got a tendency to show up whenever jesus is mentioned lol. literally woke up the same minute someone in my d&d group said it in our gc and responded lol

[unborn bane]

https://tenor.com/view/jesus-peeking-i-see-you-guilty-gif-25117299

[fresh dawn]

noooo

resend it

I need to save it

this is why I store deleted messages on PC discord

saved

[umbral wadi]

(Wouldn't want anyone to think I was actually saying that)

[fresh dawn]

oh I fully intend to use it legitimately

[umbral wadi]

I'm sure but with your name it gets funnier. I get trolls trying to play D-n-D Ruleslawyer on anything I say. 😦

So I have to be careful.

[fresh dawn]

I'm omw to d&d rn, and everyone in the party is a rules lawyer

it's half the fun to us

coz the dm does it at us, and we do it back at him, and to each other

[umbral wadi]

I found my old sh--posting folder! 😄

[fresh dawn]

I keep all of my memes in the exact same folder

[umbral wadi]

Including old memes I made myself.

(Sadly I still have a huge music video directory to sort out still but who has time when you're making music and music videos now?)

[fresh dawn]

I do, because same

[umbral wadi]

Since we're coming up on it soonish, here's this old gem....

[fresh dawn]

how everything in my collection is stored lol

[fresh dawn]

i got one of the christmas cat ones for a friend last christmas

[umbral wadi]

This goes on way too long. But that's why I grabbed it. I present to you: Who Asked?

(As an aside, look at us, actually being off topic here in #🗣️┃chill-chat! Amazing...)

[pallid sable]

Hey, so have any of you guys had trouble distributing your suno songs with Sound Cloud? Like using their distribution service? They keep rejecting my submissions, and they won't tell me why.

[trim swallow]

hey guys do you know how its called this sounds? (The sounds were extracted from a boss theme from the Wraths Of Gods add-on of the game Terraria, the theme is called “Nameless deity Theme”.)

[obtuse coral]

hi. how can you do this?

how can you sing a famous voice?

[willow atlas]

haha... I do my best to try to find songs consistent with the singers' style, a shot of UVR5, a shot of Applio.... and Hop! The magic happens. 😁😉

[willow atlas]

Do you want more?
Come say hello to my channel and don’t hesitate to suggest cover ideas! 😉

[grave ore]

😢

[west hull]

Video nicht verfügbar
Dieses Video enthält Inhalte von UMG. Dieser Partner hat die Anzeige des Videos auf dieser Website bzw. in dieser Anwendung gesperrt.

i cant hear it

you tube disabled it

[amber wigeon]

@willow atlas Teach me. I want to make parody but the regular suno doesn't even like some of my poetry let alone parody lyrics

[fresh dawn]

he's not using suno for those

[fresh dawn]

only on the embed, as is the case with most copywrited music

[amber wigeon]

Do you know what he is using? I'd love to make parodies. I don't even care if it sounds like them singing. I just want to work with the lyrics

[fresh dawn]

most likely something like rvc

it's entirely for voice changing

talking / singing / whatever

basically you take

1. extracted vocals from song you want to cover
2. model of the voice you want to sing said song
3. run the vocals over that model to convert the voice
4. mix the new vocals into the instrumental
   bam, cover

and rvc even has a vocal extractor included, so the entire process is like, 15 minutes or so on one page

[amber wigeon]

Ah ok so that would'nt work for original lyrics outside of a song it sounds like

[fresh dawn]

correct. you require vocals to begin with

but

[amber wigeon]

I'm listening

[fresh dawn]

let's say you make a song with suno using your original lyrics you want some artist to cover -

1. create the song using suno
2. split the vocals from the instrumental
3. using the model of the voice you want covering your song, either run the extracted vocals over it, or sing it yourself (much better quality)
4. mix the new vocals with the track

coz if I'm understanding correctly, you're saying you want xyz artist/person to cover your song?

[amber wigeon]

No I'm wanting to make a song from lyrics but they are parodies of other songs, like weird al does. Sadly Suno is too strict for that considering even my original lyrics sometimes gets rejected. Although You have given me an Idea for my original stuff.

[fresh dawn]

oh I see. so custom lyrics, set to the same tune as another song?

[amber wigeon]

Is that possible? I just wanted the lyrics sung, but if I could get it to mimic the original that would be awesome

[fresh dawn]

I mean that's what a parody is

same instrumentals (roughly) with new lyrics

[amber wigeon]

Yeah I just didn't think it would be possible

[fresh dawn]

are you capable of singing?

[amber wigeon]

like a dying cat lol

[fresh dawn]

because I'd use rvc for that (assuming you wanted some artist to do it)

[fresh dawn]

free autotune exists

but yeah for now there's not really anything that can generate singing vocals set to a specific tune. you'd have to either take a long time pitch adjusting in a daw to set it appropriately, or just sing it yourself and run a basic autotune over it if you need

[amber wigeon]

Really? I guess I only looked on my phone now that I think of it. All the apps I tried claimed it but never delivered

I do have a nice microphone. Maybe I'll give it a try. Any recomendations for the autotune app?

[fresh dawn]

I'm not sure about which ones can be used inside of freeware daws

but lists like this are useful
https://bedroomproducersblog.com/2011/07/25/bpb-freeware-studio-best-free-pitch-correction-auto-tune-vstau-plugins/

[amber wigeon]

I'll have to look into this more. New to all this. Thank you for all your help. I really appreciate it.

[willow atlas]

Click on the link : Some YT videos are not playing on Discord 😉

[fresh dawn]

I saw that checkmark @willow atlas lol

[willow atlas]

Because you are right about the process 😉

[fresh dawn]

you just using the rvc extractor for vocals or something else? (didn't click them so zero clue what your quality is)

[fresh dawn]

bros just pumping them out

[tawdry fiber]

There are some "Nina Simone" classics that would be an AICover base for Amy Winehouse

Please don’t let me be misunderstood, for example.

[willow atlas]

Good idea

[solar holly]

tired of repairing one song only to achieve somewhat satisfactory audio quality compared to the original, too many repetitions to drown out the background noise.

[umbral wadi]

No Sinnerman?

[tawdry fiber]

Nina Simone has a huge list of classics. And she has incredible similarities with Amy. It's just that "don't let..." became more famous (I think) with Santa Esmeralda, and is less well known with Nina, but there are many classics: "feeling good" is also masterful.

[fresh dawn]

have this 10/10 pristine quality metalmaxxing og cover of doofyboiyo getting vocalpilled

[compact compass]

@willow atlas How do you do it mate? 😮

(Those cool covers)

[willow atlas]

RVC Applio 😉

[fresh dawn]

my process (his likely isnt that far off from it)

1. grab yt-dlp to download audio files straight from youtube (or any downloader ig, but fuck ads), may need to use ffmpeg to convert to wav
2. grab applio which comes with uvr5 embedded
3. when applio is running, head to the uvr5 section and (for ease of use, not necesarilly best) select mdx the kim vocal 2 model (you can find better model combos/chains to use here)
4. drop your wav in and hit convert
5. grab a rvc2 voice model youre after from ai hub (im not linking other discords here lmao)
6. use the downloader in the extras tab of applio to install it easier
7. head to the first window in applion, select the model and its index file, select the extracted vocals (these are all in the dropdowns)
8. hit convert or whatever it is
9. once its all finished up i launch ableton (audacity probably works just as fine) to eq out the pops in the vocals, fix where the extractor mixed up vocals / instruments, and add an fx chain to mimick the original audio (applio isnt really good at mimicking the vocal effects used in the original audio)

tweaking some of the advanced settings in applio, plus making a couple generations of it with the same settings and importing them into a daw to join them up is how you can get more powerful vocals to work out. they kinda suck if you just use one generation

in retrospect, given that all of the above tools can be ran command line. theres no excuse for me having not just made a quick tool to send a youtube link to and it just automatically build covers lmao. just need a bit of daw touchup

[earnest pelican]

Give me among us

[cosmic tartan]

What image generator does Suno ai use for its cover art, it's such a unique one

[frank panther]

I think it is SDXL...

[tawdry fiber]

Off-topic comment. It's curious how far remastering has progressed with images (videos or photos). I've seen impressive recoveries of apparently unusable videos and photos. Some solutions are even opensource. In audio/music, everything is still primitive (except what is patented). Removing noise is still primitive. Suno seems to be struggling to generate less "bad" audio. And I'm not even talking about "recovering old audio". If the 128K barrier only allows us to use Suno as a "draft", I hope there are lights at the end of the tunnel so that music with AI doesn't "die on the beach".

[fresh dawn]

same as it's.. basically always been. it's as good as the person remastering. I'm kinda confused what you're talking about since you started with images and ended with audio lol

and 128k is hands down just a current thing. it's probably due to a combination of computational requirements and current state of the chirp model.

[tawdry fiber]

Analogy

[fresh dawn]

point stands, it's as good as the person remasting (for both, audio and visual, honestly)

better you are at photoshop / image manipulation the better your result from an AI base pic will be. when it comes to audio, the better you are at sampling / building tracks the better the result

[tawdry fiber]

In photos and videos, AI has generated "final work". In sounds "drafts". It's not the same thing.

[fresh dawn]

it really is

I've worked with stable diffusion since it's inception, keeping up with the models and utilizing high end gpus with it. it's yet to produce something I personally would call final work

which mind you, "final works" and "drafts" are both subjective terms. so it changes between people

[tawdry fiber]

It depends on the prompt you use. I'm a trained designer and have been working in the field for 42 years. The limit of AI for images is canvas. But for generating the image itself, I would judge the final work to be better than many designers. It also has limits with typology

[fresh dawn]

if you're saying direct ai output is final product worthy, then canvas isn't a limit considering brush strokes can be calculated and there's already bots that paint.

but I digress, its subjective. we just disagree on what a final work is

beginning to feel like your question was less of "how far has image remasting gone" and more of "why would someone remaster it"

[tawdry fiber]

If you have to provide an illustration for a story, for example, it's the final product without retouching. If you need to supply a poster with ceilings, it's a draft.

[solar holly]

did it work?

[tawdry fiber]

No. I just wanted to differentiate between "draft" and "final art".

[fresh dawn]

your last message didn't do it

i'm aware of what the difference is. it's still subjective.

I see pretty much all creative ai output as a draft

out of the tens of thousands of images I've made with ai between different methods and dozens of models - I still see it that way. prompting is something I've always been good at, because it's just the offspring of programming and tagging, both of which I've done my entire life. so that's not the issue

[tawdry fiber]

AI for images/video is more mature than for sounds. I'm not even going to include AICover in the discussion, because I don't consider it authoritative( yet). What's not clear?

[fresh dawn]

how just needing to provide an illustration equates to final draft, or why posters require ceilings

[velvet token]

barely anyone works on top of all the AI images althought it would be needed to

i guess the same happens with music and video as well

[frank panther]

Audio has to work over time, and across multiple continuations....
Music seems simple but it's more like Animation than still images...

so the analogy is not really current AI music to old image gen...

there has been generative 'noise' and simple beats for a long time...

[fresh dawn]

I made my first ai generated music in ~2011/2012

it was bad lmao

[velvet token]

and i was an idiot back in 2011/2012

what an accomplishment

[fresh dawn]

but also AI has been used in music since like, the 70s

[velvet token]

AI doesnt equate generative AI thats what a lot of folks forget

[fresh dawn]

facts

[velvet token]

and also Stable Diffusion was by far not the pioneer of generative AI

not even in the modern era

[fresh dawn]

Eliza chatbot, 1960s. semi modern you has jarvis and the likes

[tawdry fiber]

The illustration (on its own) depends on a good prompt and the result can even be used (without retouching). But if the illustration is part of a context, such as a poster, book, etc., you can't (yet) automate it with a prompt. That's why I said that the illustration can be final art and a poster not yet.

[fresh dawn]

that makes a bit more sense. I still disagree. again, its subjective. I personally would not see nor utilize an illustration on its own as a final result without retouching it.

[tawdry fiber]

This is already done to a large extent to illustrate texts on the Internet. But the choice is yours. And... Managerial or not, it's just a tool.

[frank panther]

I think this is the bigger issue...
the gens need to stay consistent across an extended 'work'...
music is more like a finished animation..., not just multi-frame continuation over time, but across multiple edit/cuts and tone changes....

[velvet token]

well there is text to template by Adobe

so technically you can prompt a poster

this was made within Adobe Express with the text to template feature

i didnt touch the text at all although i can

[tawdry fiber]

Well... first you have to know to what extent the creation can be "translated" into a trained model. The purely technical part is a question of time. AI slips absurdly in understanding metrics and rhyme structure, for example. It understands theoretically, but it can't generate consistent content.

[fresh dawn]

and that's gross, there's a reason I block just about everyone I see promoting the automated chatgpt + stable diffusion combo to produce and sell books lol. do I think it's a neat concept to try? sure. to profit off of claiming it your own work? not at all.

[tawdry fiber]

Yes, I know. I use and sign Photoshop. It's my work tool.

[velvet token]

i just wanted to point out that you can actually make use ready posters 😄 Well someone more serious will not leave generated stuff untouched or unlikely

applies to music, art and co as well

or dont use those for the final product at all

but more in the process before

as a reference or idea

[tawdry fiber]

I realize that there is a group that defends the indefensible: “give human parity to AI”.

When an accountant uses a calculator, it does not become a co-accountant;

A doctor when using a stethoscope does not appoint an auxiliary doctor;

When a musician uses a microphone, he doesn’t convert it into co-singers;

Same for those who use photoshop.

Why should it be different with SD, Midjourney, ChatGPT, Suno?

I see them as tools, facilitators (even when they do most of the work).

It is one thing to imagine that some jobs will cease to exist because of the “facilitators”... It could be.

I see no problem in using tools to streamline some processes. This does not make the tool co-author, if it is your concern.

[fresh dawn]

pretty sure you're completely misunderstanding me here. I'm not against the use of ai in creation at all. I just like

• doctors to use their knowledge as well as the stethoscope to diagnose
• accountants to utilize knowledge of tax law to verify they're calculating the right thing
• musicians to understand note progression and how their vocals fit with the music

[frank panther]

Everyone thinks they will get rich on generative AI. 🙄

If it worked like that, these companies would be selling the final product and guarding the process, not selling generative AI.

[tawdry fiber]

Of course. Of course.

[velvet token]

yeah, no wonder all the NFT bros jumped in instantly

[fresh dawn]

but also yeah, many jobs will cease to exist due to ai. it's not the creative ones tho, those will end up booming. which is why I support ubi but that's an entirely different topic lol.

and nothing wrong with using creative generative ai to streamline content, at all. just taking its output and calling it final.

I'm not even concerned about ownership. I'm concerned about degradation of quality, which has become blatantly obvious from people calling their ai output final.

[velvet token]

and one can mostly easily tell people are uneducated and dont know sh*t about art, coding, music and co.

but they are the first to say that generative AI is the future and traditional tools will vanish soon

[fresh dawn]

I had 100eth when it was $6 a pop, 50btc at $100

[tawdry fiber]

When you don't use a ready-made prompt, you're using your knowledge and technique to generate something compatible. You're just replacing it with a brush.

[fresh dawn]

and that brush is on crack

[velvet token]

im not sure i understand 😄

[velvet token]

rich boy

[frank panther]

hahahaha
I was working on a 'it's a brush' analogy, but that wins.

[fresh dawn]

no I sold them each at those prices

[tawdry fiber]

I don't think it's a drug. You have mediocre brush "artists" today. You're going to have mediocre "AI artists"

[fresh dawn]

oh, no I meant the brush is on crack as in the brush is barely controllable. even tweaking every setting to the right place, even having a near perfect prompt and seed to work off of.

it's still not a final result. because you had an idea to create when you started, and it'll never fully match that. that's what post is for

[velvet token]

well thats the nature of generative AI

[fresh dawn]

yeah? I've never said otherwise

[velvet token]

lack of control even with products like Adobe Firefly and Stable Diffusion

they still offer the most of them all ofc but there is still a lot room left for control

[fresh dawn]

I mean, in the case of stable diffusion specifically, I'm at the level of control of building my own loras and model training. so it's not like I don't get how it works

[velvet token]

for music to be honest i barely tried any generative AI music product so i dont know much

actually Suno might be the first one i tried on my own

[velvet token]

dont worry, i didnt accuse you of not knowing

[fresh dawn]

sunos definitely at the top of the rankings right now. nothing open source compares, and no other services are on the same level

[tawdry fiber]

Many finished works by artists (and I'm not just talking about visual art) sometimes take a different form from what was imagined at the beginning. But that's okay... I understand your concern

[fresh dawn]

that as well. a lot of times when finding the perfect result that gets generated, an idea will shift as well, so you'll go to add that or modify that

[velvet token]

which is one reason why some artworks take weeks to get finished

[fresh dawn]

in fact that's usually what gets me in Photoshop moreso than the touchups, since I just expect those

[velvet token]

especially in companies like Riot Games

[tawdry fiber]

Yes, although I disagree with the co-authorship rules. We're back to the question of "tools". But...

[fresh dawn]

co authorship?

I've never said anything about it

your hardware, open source software, it's your product

[velvet token]

i heard some competitors offer a lot more control and customization options but i didnt try or look for myself

[velvet token]

but everyone else can take it and use it

[fresh dawn]

it's likely they do, but in terms of quality along, suno tops all

[velvet token]

without legal persecution

[frank panther]

where did you 'hear' this? 😄

[velvet token]

who O_o

[fresh dawn]

not if it has significant human modification

[velvet token]

yup

😄

[tawdry fiber]

You must not have read the rules. Especially Suno's free covers. I was referring to Suno.

[velvet token]

i generated some vector logo designs in Illustrator but i dont dare to use them as they are generated for my future "studio"

[fresh dawn]

that wasn't in response to you. but yeah, sunos software, sunos hardware, they get rights on determining ownership of the produced file.

referring back to tools, if you go to a worksite and use their hammer to make a box with their materials, it's still theft to take it home

[fresh dawn]

supreme court

[tawdry fiber]

You use photoshop, but you don't have to use "made with Photoshop"

[fresh dawn]

technically, I can't use Photoshop for commercial use for the exact same reasons one can't use suno free tier for commercial use

[tawdry fiber]

I didn't understand that part "you can't use Photoshop for commercial fibs"

[velvet token]

if you use illegal copy of Photoshop for commercial purposes you are in trouble

[fresh dawn]

waddya gonn do about it

[fresh dawn]

I don't own it.

because I have not purchased a license from Adobe, I'm not permitted to utilize their tool for commercial services. just as if you don't purchase the rights from suno, you also aren't permitted to utilize their tool for commercial services

but back to the co-authorship topic, it's not the ai that owns it, it's suno the company

[tawdry fiber]

Yes. The restriction in this case is licensing. In the case of Suno, it's complicated because there are free accounts. If they exist, it doesn't make sense to restrict commercial use or the need for "made with".

[fresh dawn]

it makes perfect sense though.. it's not really any different than trial periods other software utilizes, which oftentimes includes watermarks

[velvet token]

they simply expect those who make money with their product to pay for it, the free license is basically just a intro

and for some hobbyists

reminds me of Fusion360 by Autodesk now

althought, not the same

you can still make money with the free version i gotta look up. Edit: Ok you cant

[tawdry fiber]

Trademarks don't make the software you've created something you co-own. That's a different discussion

[fresh dawn]

trademarks were never mentioned. watermarks are the equivalent to attribution

and you don't co-own free tier suno outputs either, suno owns them

[tawdry fiber]

One of the most curious issues I've seen with these AI tools is that they claim that without violating the copyright of the artists it would be impossible to train a model, but when the model is trained a "watermark" is put on it. Is it all right to condemn copyright and be sympathetic to the right of the mark? In other words, is it capital taking the place of labor because creation is despicable but investment is sacred?

[velvet token]

or people monetizing their models on CivitAI

and even trying to copyright prompts

i dont even support copyright for AI images per default, copyright for prompts is even more ridiculous

[fresh dawn]

I'm in the ballpark of turning works into trainable data being inherently transformative in nature, thus not an infringement of copywrite. which, is a gray area. but that's where I am with it.

and don't get me wrong, I don't really hold any kind of copywrite or licensing as "sacred" beyond potential legal issues in the way I utilize it

[tawdry fiber]

I have the slight impression that some rules (made by humans) are going to bring order to "each". In fact, this movement is already beginning.

[velvet token]

what do you mean with bring order to each? 😄

[fresh dawn]

not the direct output images, no. but I guess, technically, the prompts would fall under standard copywrite laws wouldn't they? probably not though since they're technical rather than creative in nature

but also imagine filing the paperwork for a prompt, and then attempting to enforce that copywrite

[tawdry fiber]

That's not the movement I've seen in the courts. Microsoft, OpenAI, Meta... are having a hard time convincing judges of this argument. Even the super-powerful Apple has been punished for infringing on the Watch.

[tawdry fiber]

Regulation.

[velvet token]

if it was an image that was heavily worked on that would be a different case, but prompting and even in- and outpainting on top of that isnt something i consider as worthy of being copyrightable but thats my stance

[fresh dawn]

yeah that's my stance to on it

[velvet token]

oh, regulations. They have yet to come

i mean some are incoming, for example AI act in EU

[fresh dawn]

courts are loaded with people that would struggle to use a modern day mouse. I don't exactly take their input on the topic as useful.

[tawdry fiber]

Yes, I've been following

[fresh dawn]

it's transformative in nature. people think that an AI simply sees and copy's some work or modified them together, which is ridiculous if you understand the basics of it

[tawdry fiber]

Fortunately or unfortunately, rules are rules. And they affect.

[fresh dawn]

I think I've already expressed my views on rules lol

[velvet token]

i proposed a seggregation of AI art from human art on platforms from the beginning

[fresh dawn]

I don't personally care about ownership of ai created works, because

1. everything I make with ai has significant human authorship involved, to the extent that were proposals I've seen to be passed, I'm safe regardless
2. the only commercial usage I have for ai created works is nonprofit in nature

[velvet token]

actually not allowing AI art on certain platforms in the first place

[tawdry fiber]

Sense of self-preservation. Many people live in a science fiction movie in which machines will take over the world. I think that's nonsense. They will be tools and the consequences will depend on how they are used.

[fresh dawn]

pixiv

[frank panther]

Strawman

[fresh dawn]

fair

[velvet token]

understandable

as an artist and newly game developer it would be really against my interests to support some of the wishes by the AI art people

thats what it is, a matter of interests and business

[fresh dawn]

I mean. they will though. soon? probably not. with the assumption that agi→asi is possible? it's a given.

but that's also making a large assumption that agi isn't in of itself science fiction

[velvet token]

arent most of the things related to AGI a pure wish thinking and speculation?

people are listening way too much to AI gurus on YOutube

they call themselves AI experts without any deep AI knowledge to say the least

[fresh dawn]

to an extent. it's not unreasonable to see it as possible, it's a perfectly likely thing to achieve, and once achieved will be the equivalent to the discovery of fire.

it's mostly just a "when" type deal. and really like any "when" in technology, a bit of an "if"

[tawdry fiber]

From this point of view, it would be hard to imagine a super genius being employed by someone with a limited mind. It may be a silly analogy, but I'm not bothered by it. Not to that extent.

[frank panther]

UMG lawsuit against Anthropic may reign in the 'fair use' excuse.

These are corporations, charging fees for gens which are already legally sus.

Exceptions for Fair Use are intended to help individuals (and preserve free speech), not to create loopholes for companies to create competitive products in the same marketplace.

(this is essentially UMG's argument)

[tawdry fiber]

You have perfectly translated what I think. I believe that this will be the fate of AI in the courts.

[fresh dawn]

yeah I see that as a pretty dumb argument (on anthropics side). that's not fair use. I also see it as a failure in their software, to allow copywrite material through filters in the first place (assuming I understood what I just read correctly).

[tawdry fiber]

I'm 57 years old. I've seen "the best technology of all time" end up in a trash can.

[frank panther]

I think Anthropic wasn't filtering (as Suno does) to make clones/derivatives harder to produce.... Anthropic were encouraging it as a feature of the service.

[fresh dawn]

yeah that's... not very smart lol

[velvet token]

Adobe got the best out of it

used the material they had right to and from its own platform and nobody can touch em

[frank panther]

it definitely pays to hold all the cards in advance...

[velvet token]

Autodesk probably similar but since they dont have their own Stock platform or similar they have likely a deal with one of the providers

[tawdry fiber]

Well, this is an off-topic area. Also a place for speculation. It's too early to predict what will stay and what won't.

[velvet token]

yeah

which they did

[fresh dawn]

I'm 26, young and dumb. but pretty much spent my entire life studying technology and specifically software intelligence.

still rolling with the assumption agi→asi isn't science fiction, it will be globally life changing.

[tawdry fiber]

I've seen this movie before. A déjà vu

[fresh dawn]

tho currently the best technology of all time is my vape coz it delivers enough nicotine to keep me kicking

[fresh dawn]

don't get me wrong, I don't think terminator would ever happen lol. just.. life changing. dunno what that entails

[tawdry fiber]

I don't smoke, so it's hard to imagine it that way. But I think AI has already changed the course of humanity today, regardless of AGI. Whatever comes is profit.

[frank panther]

by analogy... airplanes are a globe-changing technology, transforming every industry...

But the average person does not directly use an airplane in their daily lives...

So it may be hard to picture what will happen, or how life will be different (other than something vague and futuristic)

[fresh dawn]

whatever comes is profit
I hope this doesn't imply whatever ai brings will be good

[tawdry fiber]

No. AI doesn't preach "profit sharing", as far as I know. But some of the rules of the game will change. I think the Berne Convention will be updated... Or not. Hehehe

[velvet token]

some of the technology simply wont be available or affordable for a lot of folks

[fresh dawn]

hopefully so will the legislation over policing, insuring, hiring, renting, etc

[velvet token]

like some Tesla AI bot in the household

[frank panther]

right, like we do not need an airplane in every home... And owning an airplane is quite expensive... but regular people do have access to airplane technology... if not for transporting their bodies, then for shipping, rescue, commerce, yadda yadda...

[fresh dawn]

about $2k

[velvet token]

yeah, i mean a bunch of us in the western world can afford some of the AI luxury stuff

[fresh dawn]

actually I'm pretty sure ultralights and paragliders are cheaper if you wanna count those

[frank panther]

not really. lol

[fresh dawn]

then yeah about 2k lol

[frank panther]

(cheaper yes)

[fresh dawn]

tbf I'm not even sure about a "bunch of us". lots of ai content comes from open source software, so the purchase is in the hardware which.. yeah I guess a bunch of us can easily obtain. but not necessarily for the purpose of ai

[velvet token]

i mean i didnt get into the details

but the average man living by the western standards can afford a lot

some things ofc only after saving some money but still

[frank panther]

their are also submarines that go to the Titanic... 😔
there's a practical difference between richman's hobby and a society-transforming technology.

Humans have been landing on the moon for 50yrs... hard to say how that has impacted day-to-day society.

[tawdry fiber]

Sometimes the problem isn't even the law, but its interpretation and applicability.

[fresh dawn]

only 9 make it back tho

[frank panther]

that would be an interesting premise...

10 people can go to Mars, 3 will be able to return alive....

I am sure there would be plenty of people signing up for that....

[fresh dawn]

10 people go to Mars, 9 people make it back is a better comparison. id absolutely sign up assuming I was signing up for the vessels designed and engineered for doing so.

[frank panther]

Yeah, I made it worse to say people would still go for it, even though odds are they would be ded.

'people'

my strawman

[fresh dawn]

I was just counting a single person as a single vessel tbf

[frank panther]

unfair of me to compare to Oceangate... there are MANY people who said no..., and one casualty was 18 there with his dad.... 😞

[tawdry fiber]

I get "hives" when I hear this applied to AI, many using webui for Stable Diffusion because it is "open source". Open nodes impose serious restrictions on the derivation of models. I use Draw Things which, in my opinion, is the state of the art for SD. But I will say that it is Open Source.

[harsh cloud]

Hey guys, i just wanted to tell you that i'm going to open a new YouTube Channel soon, wich is like NCS but for AI Music, so if anyone wants to send me their Suno made songs, I'm open to it, credits go all to the creator ofc! 😄

[tawdry fiber]

If your intention is to pay for the channel and you're using free music, forget it. I have a channel with Suno music just for recording lyrics, nothing paid for

[harsh cloud]

nah, it's just a free fun project dw haha, just to give a little plattform for AI Music

[velvet token]

Drawthings.ai? The "pocket image generator"?

[tawdry fiber]

Yes, but there's a desktop version too (OSX)

Returning to the analogy with tools... Companies shouldn't own creation either. It's as if record companies owned the work of their players... It's not like that.

[fresh dawn]

draw things isn't open source (unless I'm missing something). it's closed source, and looks to be more limited than most alternatives. also very apple specific.

I also don't see how open nodes applies, I can't find anything in relation to it and stable diffusion?

using a webui is beneficial in every way, because I can use it from anywhere through my intranet, and have full control over.. literally everything with zero restrictions.

[tawdry fiber]

"without any restrictions" even though it's just a skin for a proorietary model from a "consortium" (CompVis, Runway, and Stability AI, with support from EleutherAI and LION)

[fresh dawn]

who said I used base models? I've pretty specifically explained I use and make custom models. I should add - draw things is absolutely no different in this aspect

except that you can't make your own with it

[tawdry fiber]

In Brazil we often use the phrase: "Peru is for dying beforehand". We're at the time when they throw alpiste to the pigeons. I'm going to eat it too. But with my suspicion meter on.

[frank panther]

Draw Things is NOT basic, lol. It's the fastest SD client on Apple Silicone.
It has built-in LoRA training, and is written in Swift (not python)

Aspects of the technology used are open source (metal flashattention) and it's free, but I don't think it's open source...

[fresh dawn]

I didn't say it was basic, I said it was more limited.

[frank panther]

you said it couldn't train. It can.

[fresh dawn]

loras, yes

[tawdry fiber]

It's not. WebUI isn't either. And the problem isn't using customized templates, but using SD, which isn't opensource. Even if you say it's

[frank panther]

yes.

[fresh dawn]

stable diffusion is open source, the model is proprietary. there are quite a few open sourced model alternatives though. but good luck getting a massive community to migrate to them.

the issues of restrictions coming from that, are almost immediately circumvented once custom merges and weights are integrated to the base model.

[tawdry fiber]

No problem. Let's be happy with what we believe in. Using an analogy, I find the "Harmony OS" style of being more coherent than "An opensource Android" full of limitations. If you know what I mean.

Opensource ... Just not

[fresh dawn]

custom roms >>

not really a difference in the issues between android and harmony os

[tawdry fiber]

To use a hyperbole: "it's to free yourself from the chains of Google"

[fresh dawn]

oh I use the internet. that'll never happen

[frank panther]

Me every few weeks: "There has to be an alternative to Youtube... Oh gawd, they are even worse somehow..."

[tawdry fiber]

I think that's one of the feelings that should be generated when I say "opensource". Haweii has an Os to call its own (regardless of whether it's worse, the same or better). AI that depends on a "matrix" to work bothers me.

[fresh dawn]

but matrices are very important for the algorithms behind ml ||this is a joke||

[tawdry fiber]

Linux has a controlled kernel, but it fits more with what I imagine as OpenSource. To be clear, I'm not proposing "technological anarchy"

In fact, practically everything today is derived from Unix

[tawdry fiber]

Does anyone who lives in the countries mentioned know if this information is true? "... In the United States, copyright rules deny the existence of a "non-human" author, for example. The UK legal system considers that the right to works created by artificial intelligence systems belongs to the person who made all the necessary arrangements for the creation of the final product. In Portugal, works created by AI are considered to be in the public domain."

[tawdry fiber]

It seems that the only known case of AI authorship was in a Chinese Court in 2020 by Tencent and its Dreamwriter software.

[frank panther]

In the US, only a human can be an author. US Copyright Office has plenty of experience dividing a work into its copyrightable and non-copyrightable parts (it is not either/or – it is lyrics by human, music by AI, images by another AI... etc)

Cherrypicking a US case that seems to be an exception but isn't, a single piece of computer art from the 1960s was (eventually) granted copyright when the author proved he'd written every line of code in the algorithm
(A. Michael Noll “Gaussian-Quadratic”) – his later art was denied copyright as computers evolved past the possibility a single programmer.

[frank panther]

as a technicality, the work was not deemed to be 'made by AI', but by a human creator (assuming this is the case linked)

https://www.lexology.com/library/detail.aspx?g=eee49487-a815-404c-8e20-2c8a702ebb63

(China is not exactly the leading authority on international copyright, and the case was not considered to be an actual ruling on copyright)

[tawdry fiber]

Thank you!

I like Portugal's model of considering the public domain

[frank panther]

yeah... as i understand US copyright... 'public domain' isn't really a thing in US courts..., it's a word for a copyright in limbo...

we can't make something to be legally 'public domain', it's a sort of 'until proven otherwise' kind of 'non-status' designation... (that's not how regular people think of it, oc)

The movie 'It's a Wonderful Life was assumed to be in public domain, until someone got some lawyers and put a stop to it, lol 🇺🇸
https://en.wikipedia.org/wiki/It's_a_Wonderful_Life#Ownership_and_copyright_issues

[tawdry fiber]

This is thorny terrain. The agreement reached between the studios and the dubbing union about synthesizing voices is reckless. On the other hand, Madonna's decision to forbid the use of her voice, music and image in her will even after her death reopens the discussion about "moral rights", which are more complex and subjective than copyright

[frank panther]

🥺 I didn't hear about the Madonna thing....

[tawdry fiber]

I'm a fan of this movie and James Stewart

[frank panther]

what is reckless...? (I haven't caught up on this story)
I thought it is for generative dialog (?) and requires a special contract about creating the custom voice model and how it's used....

(although in the context of lapsed copyright... 🤔 Hmmmmm all those voices will be public domain someday, even Madonna's)

[tawdry fiber]

The Berne Convention (Article 6bis, if I'm not mistaken) talks about moral rights, which are different from copyright. If the author has expressly made this prohibition and life, even after the copyright term has expired there will be a restriction

Copyright is more about being paid for use. Moral rights have to do with whether a work has been defaced. It's complicated but not impossible to prove. With the prohibition in a will, this is more possible.

This article talks about her banning holograms: https://www.culledculture.com/madonna-holograms-billboard-music-awards/?ssp=1&darkschemeovr=1&setlang=pt-br&cc=BR&safesearch=moderate

https://www.thesun.co.uk/tvandshowbiz/22988839/madonna-strict-rules-hologram-legacy/

[radiant sierra]

Welcome! For help, check out Suno Help

[tawdry fiber]

the part where she talks about music and voice was for an article on brazilian tv. I couldn't find the link.

[frank panther]

🙏 📰 👀

[solar holly]

My journal today with Suno

1. Do not follow the given prompt but repeat the previous prompt in the second part. (Click to continue.)

2. Generate a song that is partly according to the prompt and partly unknown in what language is used.

3. Make an intro or background song without vocals.

4. Generate a song in an unknown language and end the song at the end of the prompt.

5. Repeat the four points above. even though I have spent more credit.

[quartz ridge]

https://tenor.com/view/foekoe-foekoe-gaming-the-greenscreen-room-cats-kitty-gif-21035271

when the suno ai refuses to sing lyrics

[subtle hearth]

Tru or refuses to gen past part 8

[fresh dawn]

only just now messing with sdxl, dont think ill be able to really
this is for a single iamge

[compact compass]

@fresh dawn Really cool. How you did Plankton voice?

[fresh dawn]

#🗣️┃chill-chat message

the real question is how I predicted the future

[tawdry fiber]

We are in the age of dazzle. When many who are unfamiliar with a paintbrush produce beautiful paintings. Those who don't know what rhyme structure and meter mean see "their" lyrics become songs. That's nice. But I hope for the future... I don't know how far ahead AI can be used, perhaps to found a new school of "painting" or a new musical style to call its own. If it comes to that, it will do more than satisfy egos. I hope I live to see it ... p.s.: by the way, trying new styles by mixing old ones is what I've been trying.

[cosmic tartan]

how would Suno look like as a music artist? As in appearance. And realistically, no cartoons

[fresh dawn]

speaking of mixing styles. this is an ai cover of a 2 week old song (pretty scene girl, clover) about a 20 year old sense of fashion. the song being a remix of a 17 year old song (pretty rave girl, s3rl), which was a remix of a 25 year old song (daddy dj) which.. may? have been original lol ||altho in retrospect, they're all edm, just different genres of edm||

music is wild

[fresh dawn]

miku

[cosmic tartan]

I'm actually serious

[fresh dawn]

so am i

[tawdry fiber]

My previous comment wasn't directed at any specific post. It was a ramble about how AI has been used. To paraphrase Fernando Pessoa: "rambling is necessary, living is not". In fact, Pessoa himself also "paraphrased" this poem.

[drowsy osprey]

"Me and people must stay together."

"I can make you smile, you are sometimes crying"
"Forget your problems, try to smile again"
"If you really want it"
"I can really make it"
"Good enough, good enough"
"Therefore we'll try"

"Please don't stand me up, take my hand, come together"

[proud sky]

can you generate high fidelity good audio songs? They all seem to have a hiss which is really anoying especially near the last fiew iterations

[tawdry fiber]

No. In the web version, where the quality is better, 128k. The quality is still very poor. It's good for drafts. P.S.: bitrate is only one of the problems.

[solar holly]

very bad experience tonight, none of the songs fit, all stuck on Part 2, check in the library wasted just like that 8 pages.

[west hull]

ya sometimes ou have bad days

[primal needle]

hello can i talk to someone

[radiant sierra]

Welcome! For help, check out Suno Help

[languid robin]

МИСАК КАРЕН АРТУР

[radiant sierra]

Welcome! For help, check out Suno Help

[grim idol]

Pretty interesting sampling technique in MAGNet model https://arxiv.org/pdf/2401.04577.pdf ! I tried to do something a tiny bit like this in Bark, but by revising/editing generated sequences, the facebook method is wild

[fresh dawn]

i thought this was really good
until i searched deeper and found the third eye awoken

also hairband, rip

[velvet token]

Id inpaint it

[fresh dawn]

eh. not a huge fan of inpainting. i redid the upscale, and like that results head much better. gonna take them both into photoshop and merge stuff up then touch up whats left

actually got the heterochromia base too on this upscale

[velvet token]

That works as well

I work almost always in Photoshop

[fresh dawn]

tho needs a bit of adjustment in scalera shape

same

[velvet token]

Wait what do you even use with that img. Stable Diffusion?

[fresh dawn]

yeh

[velvet token]

Aah okay

[fresh dawn]

its a 1.5 based model, sdxl hates me lol

[velvet token]

SDXL is nice but ressource heavy and therefore not ideal at all to work with for me lol

Until i get the next generation of RTX

[fresh dawn]

same. i need a new gpu for sdxl to really fit in my workflow. might go ahead and load up the model though and see how it does with this upscale, could yield neat results

[velvet token]

1.5 is not as bloating right?

[fresh dawn]

not at all. relatively small model sizes and I make 3 1k images in ~5 minutes on a 1070

then the upscale to 2k takes about 8 minutes

[velvet token]

Oh okay 😄

I used SD only for like 2 weeks

SDXL

[fresh dawn]

been using SD itself for over a year easily. started sdxl a week ago ish

[velvet token]

I remember the Disco Diffusion times still

Crayion 0_0

Then i went in for Midjourney in july 2022 up until december that year

[fresh dawn]

I remember waifu2x releasing

[velvet token]

0_0

[fresh dawn]

oh shit

[velvet token]

I need to play around with the Alpaca plugin in Photoshop

[fresh dawn]

we have the most niche mutual server I've seen here so far

[velvet token]

I completelly forgot about it

[velvet token]

Project Owl? 😄

[fresh dawn]

for sure

[velvet token]

XD

[fresh dawn]

got a tracelabs invite yesterday

[velvet token]

Whats tracelabs

[fresh dawn]

https://www.tracelabs.org/

got an invite for a ctf team tho

[velvet token]

Oooh 0_0

I used to spy and sabotage neonazi networks

But not out there but in social media

[fresh dawn]

that's always a good time

bought my first ml rig by blackmailing ah.. people worse than nazis

[velvet token]

[weary orbit]

Its just me that get mad that Suno deny a original lyric by saying its not original?

[fresh dawn]

many people do, but it's just a false positive / bug #💬┃general-chat message

[weary orbit]

Thanks for the info

will keep in mind

[fresh dawn]

usually even adding spaces in certain places or extra lines can fix it, sometimes phonetic spelling of the word triggering the error

[tawdry fiber]

They are checks and balances. Self-protection until there are homogeneous positions (or close to it) on intellectual property using AI. I see it as normal…Or simply bug

[fresh dawn]

gettin somewhere at least, kinda

[tawdry fiber]

Lately I've been getting the impression that the songs generated are not very creative. I've even used lyrics made by AI and those of some little-known poets. I've used styles I've never tried before, but I feel there's little variety. I don't know if it's just me. Even the songs posted on Discord.

[umbral wadi]

Your first post is to try to ping everyone a video?

[soft plover]

Hi friend

Why so mad?

[umbral wadi]

Please read the rules of the Discord. One of which is rule #5️⃣.

Yes this is #🗣️┃chill-chat , but this is offtopic for even #🗣️┃chill-chat !

[soft plover]

I followed the rules.

Do you?

[umbral wadi]

I just don't believe we're looking for politics here.

[tawdry fiber]

That's perfect. I think this topic is too "off-topic". If moderation doesn't start deleting this type of post and the countless spams, it's going to become a mess.

[umbral wadi]

Precisely my concern. The internet is a large place and so is Discord. There are plenty of other places for that discussion. This does not have anything to do with music or with Suno or even AI...

[frank panther]

I've removed your spam.

If you do it again, you will be getting a time out.

[soft plover]

I will ddos this server.

Dont censor me.

Want me to go my Pegasus files?

Do it just for you.

[frank panther]

Don't tell me what to do.

Time out for 24hrs.
Thanks for playing.

[umbral wadi]

I dunno...threats to DDOS? That needs to be more serious I suspect.

[tawdry fiber]

This is a cowardly act. No one is against discussing the topic. But what's the point of posting here? It has nothing to do with censorship.

[frank panther]

Sounds like a threat. feel free to ban.

[umbral wadi]

@tehsystem has been banned and this threat will be reported to Discord admins as well.

[fresh dawn]

lol that was a fun read

1. can't ddos a discord server, unless you get into the discord servers themselves to determine what node in what data center that server is hosted in (as well as the backups)
2. pegasus is a government utilized / developed malware for civilian surveillance on mobile devices, code isn't available (except the backdoor itself)

I think lil bro up there recently discovered netsec.

[mossy garnet]

Literally what I was thinking, is he going to DDOS discord itself, or?

[fresh dawn]

bro must have a mighty hefty warehouse

[hearty condor]

Hey! I need a bit of help here. I'm trying to find an expertise in audio generation, and cannot even understand how to properly search for such help. Bark is quite slow, and the output quality isn't great. I found that I can get better results by training my own RVC model and applying it on top of Bark output. However, even with that the quality of RVC sometimes is extremely poor. I'd like to be able to generate longer audio, let the audio narrative semantics match the context, and improve on the audio quality.

At this point I feel like I need some expertise -- a person, who understand the field, and would help me. Would you have any references, or at least an idea about how to search for such an expertise.

[grim idol]

Check the history of channel bark-technical (Or my posts in there) for some pointers. But there isn't a lot of info out there for Bark reallly

[tawdry fiber]

Soft, sensual voice with good vocal nuances. Lucky or prompt?

[fresh dawn]

neither it's a vocal replacement

[tawdry fiber]

Do you know of any project that proposes to do the opposite of RVC? With an "a capella" generating the instrument automatically?

[fresh dawn]

I'm confused because that's not the opposite of rvc

[fresh dawn]

so this is the original
https://youtu.be/s1tAYmMjLdY

[tawdry fiber]

The opposite is an exaggeration since the RVC needs a voice to be cloned. I'll rephrase the question. Do you know of anything that automatically sets the instruments to "a capella"?

[fresh dawn]

I'm still confused what you want. like something that replaces instruments with other instruments?

or something where you input a voice and it automatically creates an instrumental for it? (because neither, well, kinda some Ableton tools)

@grim idol if you're up

[tawdry fiber]

No. Something that creates instrumentals where there's only the voice (singing, obviously). From a prompt.

[fresh dawn]

probably not

[tawdry fiber]

I imagined it would be technologically simpler than creating voice+melody, since the voice already has sound cues.

[fresh dawn]

eh. rather than starting from a point and moving forward, it's have to be consistently based on a small sample set

for ex the tools I mentioned in Ableton let you extract the melody and rhythm from an audio clip, you can just add some midi instruments to that to have anything you want.

ig it's just technologically more simple, to the point that ai for it is overkill

[tawdry fiber]

Good. I think it would be an evolution for those who sing with pre-programmed media accompanied by Yamaha keyboards, for example.

And some tools are quite technical for ordinary users.

I think one of the functions of AI is to "make it easier". "Translating" for simple use what is often used by a "niche" of IT people.

[tawdry fiber]

https://inboom.ai/microsoft-ceo-rabbit-r1-ai-gadget-was-most-impressive-demo-since-steve-jobs-iphone-unveiling/

Something very off-topic. Did anyone follow the announcement of the Rabbit R1? I found the proposal promising.

[umbral wadi]

I wasn't impressed.

Here is my mini-review I wrote after someone in my own Discord posted about it and I finished watching the video on their webpage:

It's a solution in search of a problem.

The biggest blocker is that it is competing against the inertia of the smartphone industry. And it still needs a massive infrastructure to pull off what it does. Apple could disapear tomorrow and Google the day after and most of my Android stuff could continue to work for quite some time. The iOS stuff less. This thing? If there servers go down tomorrow you have a very nice looking orange brick.

I have no idea what the actual lag time on stuff does to real world use of the device. And while it is interesting that you can train it the way they demonstrated, that also means you're being given a less smart device and creating value by training it for the company. Most of the applications demonstrated seem to be more of value as a business class device and less as a casual or non-business user. So at what point can it offer entertainment options (other than music) or access to a digital assistant to talk to?

The one thing that it showed with its ability to make a meal from observed items in a refrigerator is intriguing, but webapps have been around for a while that can do this for you, you just need to manually enter in the items you have to work with. That seemed to be about the be-all and end-all for a causual user. How often does this guy think people who are not CEOs take trips like this? That feature would be used, what? once or twice a year at most?

Overall the software is interesting, I expect that to continue to grow and its use as an application to be bought and sublimated to something like Siri or Alexa a benefit for those who use them. The hardware is much less interesting. The need for perfect connectivity less so. The issue of everyone constantly speaking to their smart devices, significantly less so.

It's a toy.
Kind of reminds me of the PlayDate: https://play.date/
That said, he's absolutely correct--we are in the very beginning and early stages of what is going to be coming when LLMs get added to pocket computing. He's made a bet that things will go a certain way. He could be absolutely right and his R1 companion could sweep the market. I just don't find it compelling--yet.

Now if someone has some additional information not in the CEO video where we see something compelling, I'm interested to learn more.

[quasi brook]

Just sub to your channel

https://www.reverbnation.com/tracylee1/song/34321279-unbreakable-love

I want to get right promotion of background singers and me singing the lead when I can get to a room that is not so noisy to record I will work on that project

[tawdry fiber]

But it seems to me that the model is hosted on the gadget itself and not on the Internet, and this could be one of its distinguishing features. What I like is that a new idea has finally emerged that could replace an application to perform a small task. As it would operate in the background, it would be possible to interact with the operating system in the future without the need to use a separate application for each task.

[umbral wadi]

The problem is it's not replacing apps. It's facilitating them.

Or to put it another way, it's filtering the information from those apps. So you're not free from the apps, you're simply adding a middle man. And I'm not sure that it makes sense to do that.

[tawdry fiber]

not at the moment, but what I understood from the release was that the next phase would be exactly interacting with the system kernel and not with the applications

[umbral wadi]

Also I go back to my original point--what does it do for the average or casual user? How often do you travel? So how often will you use its biggest feature?

And once you remove simple communication, once you remove the ability to play games and entertainment from the mobile device, what's left? Taking a picture of your refrigerator to build a recipe for dinner? I can probably find five websites that can do the same thing. Without the added delay of taking and processing a picture first.

It's a solution seeking a problem to solve and I just don't see the problem.

[tawdry fiber]

It's an Alexa that doesn't depend on subscriptions or the internet.

[umbral wadi]

Have they said it was local? I haven't seen that anywhere.

I'm really not trying to be negative. I just don't see it.

[tawdry fiber]

Yes. Ease is the difference

[umbral wadi]

Maybe. It could be you're right.

But as someone who has never had great luck with text to speak I'm doubtful.

And how long are they going to get away with being a middleman before someone starts demanding rent?

These are factors that people need to consider.

I wish them the best and I hope genuinely to be wrong. I'd love to be hearing about the Rabbit R2 device in a year or two that does all that and fresh toast.

[tawdry fiber]

he even compares it to another gadget that was presented by former Apple employees but which depends on ChatGPT internet and subscription. according to him the rabbit does not

[umbral wadi]

Let me pull up the video again. What time stamp, do you remember?

I'm scanning the transcript and I'm not seeing that reference.

They claim speed over other voice AIs but not that it is local.

[tawdry fiber]

24:24

[umbral wadi]

That talks about price, not that it is local AI.

Even if the AI the LAM runs the system is on Rabbit's servers, it's still dependant on a network.

[tawdry fiber]

Yes. Watch the next part “no subscription”

Have you seen the hardware specifications?

[umbral wadi]

Yes. No subscription to ChatGPT. Which means that you're getting access to something on their servers, not necessarily OpenAI's servers.

That does not mean the AI is running on the phone or mobile device itself.

[umbral wadi]

IIRC it's Mediatek, right?

[tawdry fiber]

4gb/128gb

[umbral wadi]

https://www.theverge.com/2024/1/18/24042688/rabbit-r1-ai-gadget-handheld-news-updates-storystream

So it doesn't come with one yet.

And only for a year.

[tawdry fiber]

There is storage to support the model locally. I think you can also use perplexity.ai, but not as a dependency on it, but as an option, just like on the desktop.

40,000 units sold in 8 days...

[umbral wadi]

Not shipping until Easter.

[frank panther]

literally no one stands in front of the refrigerator trying to come up with a dinner for 4 guests made from coldcuts, table scraps, and old cheese.
😂

It's 100% 'Housewife Helper' silly marketing from people who do not live normal lives. It is straight from the 1940s.

https://www.youtube.com/watch?v=DUSS6Q1sp2g

[tawdry fiber]

Well, for some reason they're selling a lot. And not the ho it is because it looks like a tamagotchi.

[umbral wadi]

I thought it was because people were confusing it with a pokedex?

🤣

[tawdry fiber]

😂

[umbral wadi]

I mean it though. I do wish them the best. I hope they make it and we're talking about the Rabbit R7 some day.

[tawdry fiber]

I hope the next AI revolution is about breaking down language barriers. And it shouldn't even be the task of an app. The operating system itself should support it. Especially in relationship apps (social networks, meet, etc). If someone on one side of the screen speaks language A and someone on the other side speaks language B, AI should work in the background so that A receives messages in language A and B in language B in real time. I'm not talking about a translation bot, but a kind of reverse Babel. Using English as an intermediary is a stumbling block for technology. What do you think, Discord?

[west hull]

k

[fresh dawn]

I think there's a lot of tasks out there people want to point ai at, when it's entirely unnecessary to do so. even to the extent of privacy issues and degradation of quality.

[fresh dawn]

also here's my tamagotchi

[nimble temple]

Hola

Soy nuevo en discordar y estoy curioso de lo que se trata. Hablo tanto inglés como español, pero prefiero español si hay alguien aquí que hable mi idioma.

[gleaming condor]

Hi does anyone have a discount code?

[viral flame]

There are currently no available discount codes that I know of.

[fresh dawn]

dropping this here solely to be off topic, carry on

[pale kindle]

hi

[umbral wadi]

Is this a personification of Suno Hallucinations?

[fresh dawn]

statistically speaking with how many people are in this server, 387 feel the same about that image as me. taking into consideration the discrepancy when it comes to the the tech industry, the numbers closer to 1,597

[violet notch]

ello

[quartz ridge]

this is how chat gpt thinks the history of suno AI is:

Suno AI is an artificial intelligence system that can generate music in various genres, languages, and styles. It has evolved over time, from a rudimentary prototype to a sophisticated and versatile tool. This essay will trace the development of Suno AI from version 0.0 to version 6.0, highlighting the main features and challenges of each version.
Suno 0.0 was the first attempt to create a music generator using deep learning. It was very distorted, and the only intelligible thing occasionally may be the voice, which also faded in and out. The music was not coherent, and the instruments were not well-defined. The voice was robotic and monotone, and the lyrics were nonsensical. Suno 0.0 was a proof of concept, but it was far from being a usable product.
Suno 1.0 was a significant improvement over Suno 0.0. It was mostly unable to generate subgenres, but it easily generated popular genres, such as pop, rock, and hip-hop. It was limited to English and Spanish, the two most widely spoken languages in the world. The instruments were more coherent, but still not very realistic. The voice was still robotic, and occasionally had no voice when lyrics were prompted. Suno 1.0 was a basic music generator, but it still lacked diversity and quality.
Suno 2.0 was able to generate most subgenres, but it failed to generate niche ones, such as ambient, metal, or classical. It generated popular genres like a breeze, and it could generate in 15-20 languages, covering most of the major regions of the world. The voice was always provided, but it could sometimes be robotic, especially for languages that were not well-represented in the training data. Suno 2.0 was a more advanced music generator, but it still had limitations and inconsistencies.

Suno 3.0 was able to generate all subgenres, but it may struggle with generating niche subgenres due to minimal training data in those, albeit generating them. It generated all genres, from jazz to techno, and it could generate in 50 languages, including some rare and endangered ones. The voice was always coherent and the instruments were easily intelligible. However, Suno 3.0 had a major flaw: it failed with metatags and manually read the metatags with voice (i.e. [Chorus], [Beat Drop], [Verse]), singing them as if they were part of the lyrics, instead of creating the according musical sound, causing generation issues. Suno 3.0 was a comprehensive music generator, but it still had a bug that needed to be fixed.
Suno 4.0 fixed the metatag issue and no longer struggled with niche subgenres. It could generate in over 1000 languages and also accepted sound effect and voice metatags, such as [Whisper], [Scream], or [Laugh]. Suno 4.0 was able to create music that matched the mood and tone of the lyrics, and it could also add some personality and emotion to the voice. Suno 4.0 was a versatile and expressive music generator, but it still had room for improvement.
Suno 5.0 added a variety of filters to add a feeling to the song (i.e. distorted tape, clean, unpolished). It could also be provided clips of other songs for reference and make the song based on that. It could continue from clips and stitch together clips to make a full song. It supported over 8000 languages, including some fictional and extinct ones. Suno 5.0 was able to create music that was original and inspired, and it could also mimic and remix existing songs. Suno 5.0 was a creative and innovative music generator, but it still had one more feature to add.

Suno 6.0 added an AI-generated music video with customizable visuals that could be generated alongside the song. The music video could match the theme and style of the song, and it could also include some elements from the reference clips. The visuals could be edited and adjusted by the user, such as the color, the speed, or the effects. Suno 6.0 was able to create music and video that were synchronized and harmonious, and it could also offer some artistic and aesthetic choices to the user. Suno 6.0 was the ultimate music generator, and it was the culmination of Suno AI’s progression.
In conclusion, Suno AI is a remarkable system that can generate music in various genres, languages, and styles. It has evolved over time, from a distorted and unintelligible prototype to a sophisticated and versatile tool. It has overcome many challenges and limitations, and it has added many features and enhancements. It has become a source of entertainment and inspiration for many people, and it has also contributed to the preservation and promotion of cultural diversity and musical creativity. Suno AI is a testament to the power and potential of artificial intelligence.

[tawdry fiber]

Is this a personification of ChatGPT Hallucinations? P.S.: That's a paraphrase

[quartz ridge]

ye

[umbral wadi]

I want to live in this world...Suno v6.0 sounds like a dream come true...

[quartz ridge]

Fr

[umbral wadi]

I wonder if it'd be something that would need BBCode type syntax or something to make it work though?

[fresh dawn]

I want lyric prompts that allows you to weight words or phrases for emphases, like i am a (man:0.5) of constant (sorrow:1.3) would create a decrease in emphasis on man and increase on sorrow creating a bettrr fpow

maybe ["even tarantulas aren't immune from an ambush":spoken:rise] ["the centipede is a predator":bassdrop]

and now to listen to those songs in succession

[quartz ridge]

How come Suno always makes a picture of a bird whenever there’s no lyrics or only meta tags

[umbral wadi]

Probably influenced by the name of the model. Chirp.

[fresh dawn]

they probably utilize a primer image, that bird is probably it

iirc it's basically the same bird across the board isn't it?

[dim stump]

Hey, can someone explain something to me?

[fresh dawn]

you rang?

[dim stump]

I wanna buy the Pro Plan, it says I get 2500 credits/month, but the free version gives me 1500 credits/month (50 credits/day), so, if I buy Pro Plan do I get 2500 credits + 1500 credits (free version) or only 2500 credits, so only 1000 credits more than the free version?

[umbral wadi]

Please do not spam the Discord with the same question in multiple sections. Take ten minutes timeout to read the #📜┃rules, thank you.

[fresh dawn]

the absolute state of voice ai lol

i figured id have to record and translate to make it good, but just straight up tts

[tawdry fiber]

I have the impression that in the future it will be possible to modify an existing voice to generate a completely new voice from a model. For example: use Frank Sinatra's voice as a base and polish it (if it's possible to improve what is already perfect) by enabling some nuances in the timbre and generating a totally new voice. I don't know, but it could be the way for AI to circumvent copyright restrictions - after all, similar voices exist all the time...

[grim idol]

You can do that in Bark now, that's how I tweak and build lots of voices. Though I don't use voice clones or real voices, so it's more generic, like "angry stuttering voice + smooth talker" to make a variant.

I haven't used RVC but there must be a model merge of some kind, even if hacked on

[tawdry fiber]

I couldn't test bark on Discord (it wasn't working) and my desktop is under maintenance. I can't do a physical installation. Do you recommend any online possibilities for testing using a cell phone?

[grim idol]

There's some huggingface spaces online at times, but not sure off the top of my head of a link.

[tawdry fiber]

Ok

[fresh dawn]

[quartz ridge]

“Like if you agree” - Suno is mad at this new YouTube shorts comment trend

https://app.suno.ai/song/9f29c43a-a754-430b-a541-61e2ce6389c4

[dim pilot]

@winter eagle Most of earths greenery is burned from people that dont have access to heating to cook their food or stay warm, The energy crisis was solved 100 years ago, watch The Lost Century (2023). #🐣┃suno-showcase message

[dim pilot]

I asked this, and i believe it had them thinking - because after we use up our credits we start to get the free 5 a day again. It wasnt like that before i said something

I cant confirm if we get + 50 credits every day prior to using everything

[tawdry fiber]

Turn it into background music and it's all good! According to Suno, background music for posts on social media platforms like YouTube, TikTok, Instagram, Facebook, Threads, Twitch, or X is not considered commercial use. So, you can use songs generated as a free user for your posts on these platforms as long as you give attribution to Suno. 😊

Source

In practice... even if your "short" or YouTube video is paid for and you've used a song made with a free "Suno" account in the background, you're not breaking the rules. That's what you're submitting to.

[quartz ridge]

I’m not using it for commercial purposes

It’s just to express my hatred for the new comment trend on shorts that’s all

[tawdry fiber]

Got it

[quartz ridge]

Yeah I see everywhere it’s all just
“TEAM __
👇”
“WHO AGREES
👇”
etc

[sweet burrow]

ha, I kinda wanna make some pro-AI youtube shorts now about the Luddite arguments that 'grind my gears'

[west hull]

marcinek jest kochany

[celest seal]

join for unlimited credits

[umbral wadi]

Have you read our lovely #📜┃rules ? You'll never guess what rule #4️⃣ is...

You have been timed out for a week for violation of the policies you agreed to when joining the Discord. The only reason you have not been banned is because you did post it in the offtopic channel and more importantly, you only posted it once. But it was still a violation.

Note: You will still be able to use the webapp with your subscribed account. Just not the Discord.

[solar holly]

The songs I produce all have extreme frequency sounds, whether they are violet, blue, grey, pink, or white. I am not sure, and it greatly interferes with the quality or clarity of the song and the music. I have tried with audacity, but it results in the song having harmonic or sound defects. Is there anyone here who can help me learn how to mute the noise while mastering?

[fresh dawn]

https://youtu.be/h55PIIN0C0M

[dusk cloak]

https://app.suno.ai/song/bc4c5f04-a7ef-4458-9aa6-64682a6f6b70

[tawdry fiber]

🤘Suno wins over Heavy Metal Lover

[feral igloo]

anyone from Suno AI going to attend NAMM ?

[opal arrow]

Hello, I have a playlist of songs generated by AI. Please give them a listen and share your feedback with me. Thank you very much!
https://youtu.be/cHrg6ht_5ms

[dim pilot]

Those are awesome 👏

[willow atlas]

Thanks! 🙏
We're trying to change things up from what you typically find on the internet. 😉

[dim pilot]

Whats weird with youtube is i could get 1,000s of search results without altering the filter, but as soon as i change it to Sort By Most Recent the results go to ZERO with the same search term! I dont know how to find random stuff anymore

Newgrounds really has their users uploads figured out. It goes though a 5 star rating system and the new uploads are raw and unfiltered. I LOVE that

[blazing sorrel]

Hey I was wondering does Suno AI provide an API for generating songs?

[frank panther]

No API..., haven't really heard anything from Devs about one.

[willow atlas]

I'm not familiar with this platform. Is it suitable for sharing AI covers?

[umbral wadi]

This is awesome, was it Suno that was used?

[tawdry fiber]

You're "inside" the platform and, as far as I can see, you're the person who shares your AICover the most here.

Considering that this Discord channel is part of Suno's communication

[frank panther]

Not really, since music covers is what Suno does NOT do..., but

You are considerate enough to put your videos in the off-topic channel 😁

[quartz ridge]

Another nsfw invite ._.

#🐣┃chirp-beta-7 message

nvm it got deleted

[fresh dawn]

am I the only one that realizes he meant newgrounds lol

[fresh dawn]

last I knew, newgrounds was primarily for sharing animations.

[umbral wadi]

No.

But yeah, NewGrounds is was cool.

[fresh dawn]

haven't been on it in ages, but used to be the go-to for good animation

[umbral wadi]

Yep. Back in the Shockwave Flash days. I guess Apple really did have the last laugh there. I remember how everyone gave them so much crap for not having Flash back in the day...

[fresh dawn]

modern day equivalent is apple not supporting webm

[plush wedge]

hello

[dim pilot]

No its for Adobe Flash animations mostly. I just meant the way the introduction to uploads works; its one big list and you can see the new uploads, after votes it gets text colors and rated too bad and they get BLAMMED. And then theres best of the day etc.

Like compare that to Spotify, YouTube, App Stores… there simply not a page to just see like a long list of unsorted newest uploads to be surprised with. Its all curated

Sometimes i find the best music just typing gibberish and playing random songs 4 people have heard b4

[grim idol]

Soon
And lo, it was foretold
In the ancient Silicon scrolls,
A chosen one shall arise,
Bearing the sacred 'Delete Button'
For the Suno Chirp.
With it, the Chirpers shall have dominion
Over their musical creations,
The faithful shall erase
Their digital transgressions,
And their terrible songs
Shall be as fleeting
As the morning mist.

[grim idol]

As a result of my favorite silly test prompt, my Suno library is essentially just endless pages of toasters made of apples.

[still basin]

So far, this is the best I can do. I still don't understand how to use this server.

[sweet burrow]

i think most ppl use the website version (its v2 as opposed to the server's v1) and just chat/share on the server

[still basin]

Not sure I understand, but I appreciate that you answered me. ❣️

[sweet burrow]

what was it you didnt understand about the server ?

[still basin]

🥹 🥹 🥹

I just don't understand any of it. I'm not really internet savvy. I'm an older person.

[sweet burrow]

most of the ppl here skew pretty old

[still basin]

I wish I knew how to make and record music with my vocals. I only know how to use my icons to record me performing.

Did you see my videos?

https://youtu.be/W_k7bT0hPDE?si=m535Cms0YpKJzVSE

[sweet burrow]

yeah, Suno doesn't use outside vocals, though.

[still basin]

I'm always in the market to make new fans. I'm quite well known.

This is still a nice server. I like to see all your work.

[sweet burrow]

check out suno.ai, I think the website is easier to use

but basically you can prompt different styles/lyrics and it generates vocals and music based off that, no way to use or adapt existing music directly

professionally, its of limited use, unless you're looking for samples, instrumentals, or inspiration for a new composition

[still basin]

Can I put my own lyrics in?

[sweet burrow]

yes!

[still basin]

Cool!

I looked it up.
I'll try it later.

Can you do it on your phone?

[sweet burrow]

as far as i know, i havent tried though...

[still basin]

Wild you check out my video and maybe be my fan?

I have fans all over the world.

I sing karaoke. And perform to it.

[sweet burrow]

theres ppl all over the world here, and the number of styles you can try is practically unlimited

you can also translate your lyrics into other languages and have it perform in those

[still basin]

Cool! I want to do that.

Someone made me this:

I can't wait to try this myself.

[frank panther]

🔥 😂 🤤

[tawdry fiber]

It's worse when, after signing, you realize that the "hot" thing was just the label and the term you signed has no legal validity. 😹

[frank panther]

reminds me of those William Castle disclaimers
"If you DIE OF FRIGHT while watching this movie, we will pay for your funeral bill..."

[tawdry fiber]

☠️

[tawdry fiber]

🌶️

[still basin]

https://youtu.be/56QdgzbDNF0?si=k-SCyYPb5dKJeW96

[tawdry fiber]

https://www.apple.com/newsroom/2024/01/apple-announces-changes-to-ios-safari-and-the-app-store-in-the-european-union/

[frank panther]

APPLE: Govt regulations are putting you in DAAANGEEERRRR!

ME: Someone (anyone) else's web browser on ios? Hell YEAH!

I wish I lived in a pro-capitalist democracy like the EU. 😞

[tawdry fiber]

Where are you from? Feel free if you don't want to answer.

[frank panther]

new york city

[tawdry fiber]

https://youtu.be/wxLr02Dz2Sc?si=rBDYZnjXBP_cr4R7

[tawdry fiber]

https://x.com/ai_vision_verse/status/1750523124901380143?s=46

AI Verse (@AI_Vision_Verse) on X

MIDJOURNEY V6: TEXT

Utilise the ability of Midjourney's text generation with the prompt craft by me.

Typography design with text ["TEXT"] on a blank [BACKGROUND TYPE] background. --style raw

IMPORTANT:

• Use minimal words to get perfect results without rerolling

• For…

Twitter · Jan 25, 2024, 2:16 PM

Uau!

Firefly already did that!

[frank panther]

Hope you like Arial Thin in all-caps.

[tawdry fiber]

MJ or Firefly?

[frank panther]

MJ I guess, according to the thumbnail

[tawdry fiber]

Got it. Have you ever used Adobe's Firefly (now incorporated into Photoshop)?

[frank panther]

no, i hear it is pretty good?

[tawdry fiber]

Yes, it has the Adobe seal.

[tawdry fiber]

Ah, even though it's not directly related. Congratulations on the "Pulp Novel" script in the DT. We share the same admiration for this style of drawing, apparently. I've already used your script.

[subtle hearth]

Floridas trying to pass a gay law again.. You must be 16 to have social media..

[sweet burrow]

i would support enforcing "gating" 13-18 in most social media into a more restricted community

i know it won't "stop" anyone, but just because a 13 or 15 year old "can get their hands on drugs/alcohol" isn't a good excuse for making it readily available at the supermarket for them to buy

[subtle hearth]

I dont know how gating it helps with that

[tawdry fiber]

https://www.law.com/2024/01/25/google-reaches-settlement-with-singular-computing-over-ai-related-technology/?slreturn=20240026073458

It seems to have become a craze among bigtechs to "blatantly steal" other companies patents. I think 2024 will be a year of JUST fines for intellectual property infringement.

[frank panther]

I love Helm.... (same dev)

[tawdry fiber]

Can you do that?
https://apps.apple.com/br/app/suno-ai-music-generator/id6453519523

[umbral wadi]

I'm at least 90% to 99% certain that is not an official Suno product.

[frank panther]

wasn't Apple just telling us they protect us from scams...?

and that was their justification for preventing 3rd-party app stores?

[subtle hearth]

Lol

[tawdry fiber]

Very bad character developer. Total lack of professional sense.

[spiral wave]

I am 100% certain that app is NOT official suno product

I don't think any of us even know it exists....

[sweet finch]

does the same kind of fandom pushback in fandoms like books that have been turned into movies or shows exist in the mcu and the dceu which is based on comic books. what i'm saying is that ppl in fandoms where the books have been adapted into a movie or show get defensive if you haven't read the original source material. does the same exist with comic books and the adaptations of comic books?

[frank panther]

sure, but comic books aren't particularly respected as literary material, so the 'high brow' condescension is lost.

Book > Movie > Comic > Videogame

[pliant cave]

My first Suno song complete. Just a regret, i would have liked to cut the start and find no way to do it. Result is incredible. https://youtu.be/G1WMnA-JuK0?si=F3Q4bxOzaGb52sVZ

[broken cape]

.