This just killed Udio & Suno. The BEST AI Music Generator I've heard (yet)
ฝัง
- เผยแพร่เมื่อ 17 ต.ค. 2024
- This is the best AI music generator I've heard (for now).
#ai #udio #suno #aimusic #elevenlabs #agi #singularity #openai
/ 1788628171044053386
openai.com/ind...
Newsletter: aisearch.subst...
Find AI tools & jobs: ai-search.io/
Here's my equipment, in case you're wondering:
GPU: RTX 4080 amzn.to/3OCOJ8e
Mic: Shure SM7B amzn.to/3DErjt1
Secondary mic: Maono PD400x amzn.to/3Klhwvu
Audio interface: Scarlett Solo amzn.to/3qELMeu
CPU: i9 11900K amzn.to/3KmYs0b
Mouse: Logi G502 amzn.to/44e7KCF
If you found this helpful, consider supporting me here. Hopefully I can turn this from a side-hustle into a full-time thing!
ko-fi.com/aise...
![⭐ฮิตในTikTok!! ( มะมะมะหมูเด้ง MooDeng ) Ver. แดนซ์ ReMix BY [ ดีเจกิต รีมิกซ์ ]](http://i.ytimg.com/vi/4noyAoMa7bQ/mqdefault.jpg)
![[UNCUT] "บอสณวัฒน์" แฉ! แหล่งซุกเงินแก๊งบอส ขยันผิดที่ไม่กี่ปีก็เข้าคุก l คนดังนั่งเคลียร์ l17ต.ค.67](http://i.ytimg.com/vi/BL56qIfXYZM/mqdefault.jpg)
Can you tell this apart from human-made music? Let me know!
Note: I had to mute part of the clip at 24:00 due to Bruno's copyright
wow, i've been using suno since it was bark and open source on github.. This is getting wild fast... Is this accessible yet or just closed beta? I make like 50 suno songs a day. Can't wait to try it.
FYI, I'm starting to wonder if some of the synthwave music channels on youtube aren't maybe people that work for openai that have been using this music generator for the last 4 years. It would actually make a lot of sense...
There are some tell tale signs in the mixing and quality, but it's darn catchy. heh
Nope, not from generic manufactured production incl chart music of today. No damage there if all this is replaced by AI. If we talk music as an art form, I have yet to hear something as good as Stravinsky, Strawberry fields forever, Rain song, Paranoid Android, Miles Davis, Zappa, Hendrix...Mahavishnu Orchestra.. hopefully it'll replace all man made dross made for profit however, I have yet to be surprised by the musical output of these overhyped AI gadgets.
@@phibess give it 6 months. we've gone from terrible to this in 3 months lmaoooo
yeah it sounded more real than your voice.
The time is coming very soon where people will treat human made music like handmade furniture
lol good analogy
I agree. But that audience will be like 1%… so sad
Wow, that’s dark.😂😂 Good line though!💯
@@jhdesignerits depend
and if someone does acapella, youngs kids will be like, "woah, you sound like AI"
In the rap I swear I could hear it strain as it ran out of air and then take a breath a couple of times. Mind blowing. ElevenLabs has been in the voice game for years and is the best TTS around, and recently ventured into sound effects. I'm really looking forward to getting my hands on this. I'm imagining much better control in your prompts. 🤯
Well, we've had a good run, folks.
I'm looking forward to meeting our AI overlords. At least they are a wildcard; we already know where our human oligarchs will take us.
We are not in control. Humanity is its own animal.
This is inevitable. Biology is only 1 step of evolution.
So just chill out and enjoy life 💟🌌☮️
@@Recuper8maybe the oligarchs wont be so bad. better oligarchy than theocracy
There is a pattern when a new tech is introduced:
1. Shock and awe.
2. Assimulation.
When drum machines came on the scene in the 80's it was going to put all drummers out of work. It didn't.
The same happened for MIDI, Internet, Music Sharing (Nabster etc.), virtual instruments, virtual effects etc.
It's happening now with AI.
Each tech Shock and Awe period lasts about 2.5 years as it assimilates into society. Then things chill out.
Key Points:
- MIDI did not replace all the musicians.
- 90's sampled hip-hop sounds did not replace the orignal artists. Actually made them more popular.
- The Internet did not replace everyone's jobs.
- People share music now more than ever post Nabster mania.
- Virtual Instruments\Fx (via VSTs) did not stop musicians from buying guitars and keyboards.
A.I. is not going to replace all the singers. It will provide more diversity and opportunity in new areas.
There of course is a tier where there is some loss of jobs. If you are a drummer who is shattered cuz a drum machine took your duo acoustic guitar TropRock gig you kinda have the bar a bit low brother. ;-)
Another advantage of AI singers is their ability to introduce people to singers they've never heard before.
If done correctly, listeners may be inspired to seek out more works by these singers,
which ultimately helps them earn more money and gain popularity.
Furhtermore, at this point in recording technology, you can take an entire song in the same key as yours and create a single note or "drone" from it. In this way, you can make a sound that affects one at an archetypal level.
In English, "Stairway To Heaven" or "Hey Jude" can be converted into a single long note. No one would consciously know the source.
Music and technology are only moving in one direction (unless we enter another dark age). As an artist, it's an incredible playground. More colors on the palette.
😂😂😂
TV and Radio advertising will be all 90 percent AI
you already can make commercials with Udio orrrr standup comedian sessions lmao its nuts.
@@nemesisone8927 yep I’ve made commercials with Suno and 11 labs ..
I did use a guide vo in 11 labs .
Hmm nop. Brands and advertising clients want a great product. They want to have a chef cooking for them a meal, not a McDonald fast food… and this is why handmade music will still be used massively in commercial music.
Only shitty brands will use AI music on local television
@@alexkurt1101 it’s already started AI vo and Ai music on Mid level ads and radio .. I was mega skeptical but seen and heard work in my own campaigns .
Best time to unplug lol
Oh my god , I honestly think you deserve every single support. What could we do without your introductions to newest AND best AI technologies , amazing work ❤
Thank you!
I swear new AI is coming out so fast. It's hard to keep up with it.
no kidding!
It will be interesting to see the impact GPT5 has when released. If its capabilities are significantly improved there will be many new applications for AI.
And it's only gonna get faster, and faster, and faster, and faster, etc.
That was the prediction
Eventually, AI will create better AI. It may already be happening with code writing AI being so common. It's a snowball effect. I'm excited.
The Machine is staying up late at night doing lines of code
:-D that's how it gets harder better faster stronger. :-D
Funny as hell! Gave me a good chuckle! Did we just become best friends? Lmao 👊
it codes while you sleep!
That's the best I heard lmao
Lol
The jazz one is a mess melodically, BUT unlike suno and udio, this AI pays attention to song structure and repeats the chorus and verse melody !
Well I'm no listener of Jazz but my impression of Jazz is that it can sound like that. It was on point with the other genres though
@@oufukubintawell you know not everybody is like you… an idiot not even knowing what jazz is.
@@oufukubinta nah, the Jazz one was pretty bad tbh, other AI platforms also have some problems with it tho
Both Suno and Udio make consistent songs.
I agree that was a mess musically. Hurt my ears and made my eyes cross!
It sounds like they’re using individual sound sources for each instrument to achieve that higher quality end result. Which would make perfect sense. Clearly Suno and Udio are using finished mixed songs. So due to the stem separation used to isolate the different sounds, those individual sounds would of course all sound a bit compressed and noisy. Where as if they were using different instruments all recorded on their own, such as massive library of guitar loops, drum loops, acapella vocals and so on, the final result would clearly sound a lot better. And maybe this is what Eleven Labs are doing? Imagine if they teamed up with Splice?
technically speaking you could achieve this with ai and oscilators, ive already seen ai tools that can anazlise a sample and reconstruct it with oscilators and even output a ton of slight variations of the sample it reproduced...so its definitly possible
or band in a box. they have a gigantic collection that they have been building up for 30 years.
This is what I was thinking, I feel like this AI is using some sort of DAW to create stems for each instrument and vocals
@@swooshdutch4335yeah, Synplant 2 does that
Not oscillators but operators, as Synplant is FM.
can't imagine what will happen in the next 10 years
or even 2025
I'm seriously curious too. I wonder how Spotify or TH-cam is gonna handle these tools when they become mainstream. Will they allow AI generated content? Or maybe they will come up with algorithms to detect AI generated content and add a disclaimer on the detected content. Or what if they completely change their business model and charge users to upload and share their stuff since everyone is going to be a creator now. I know a big change is coming and I'm both excited and nervous.
Everyone always says that in response, "Wow, just wait until x more years!" How about we just enjoy the present for once?
@@chad_usa why would they put a disclaimer? It’s like putting a disclaimer on songs that use autotune lol, makes no sense to prohibit AI music. Music is music wether is made by violins, a synth, or AI.
@@chad_usa They will have to. It calls evolution
The absolute most powerful tool will be an AI that can take the music of an amateur musician, and turn it into a professional recording. For example you could make a shitty FL Studio song, or even just whistle a tune, and have the AI spit out a masterpiece based on your ideas.
yea but the laws for AI music rights they are tricky and well if you dont make it only via prompts it states you dont OWN anything as its not made by you.
now if you do put in your work upload it to lets say elevenlabs and the AI turns that track with your lyrics into an awesome song thats yours. important is that it has your 100% inptu btw a "shitty" song out of FL studio can already made to sound nice via AI Ozone11 mastering tool :)
@@nemesisone8927 It will have to change eventually because everyone will have to adapt and start using it. Whether they like it or not.
Yes. Actually this is what this technology should be in the first place. Like hey, AI, make this track of mine a trance song. Click!
This is what it currently does. It even skips the necessity to know FL.
What a time to be alive
yes!
Hold on to your papers mate!
Yup until the singing painting terminators killz us all :D
terrible time indeed
Aight no one can like this comment anymore as much as we agree
Thanks so much for playing the songs in their entirety :)
🎯 Key Takeaways for quick navigation:
00:00 *🤯 Introduction to the mind-blowing new AI music generator from 11 Labs that surpasses Udio and Suno*
- Mentions the previous impressive AI music generators Suno v3 and Udio
- Claims the new 11 Labs generator is even better than Udio and Suno
00:36 *🎶 Sample of a pop/rock song generated by 11 Labs AI*
- Showcases a 3-minute song with clean vocals and realistic instrumentals
- Notes it was generated from a single text prompt with no edits
04:24 *🎷 Sample of a jazz song generated by the AI*
- Highlights the emotional vocals, catchy chorus and trumpet solos
- Appreciates the clean and present vocals without a chorus effect
08:30 *💔 Sample of a smooth R&B song with electronic elements*
- Follows the prompt for BPM, musical elements like drum machine, filtered piano
- Notes the song isn't exactly the specified 104 BPM
11:57 *🎸 Sample of an indie rock song with 90s influences*
- Appreciates the combination of clean and distorted guitars, driving beats
- Slightly faster than the specified 120 BPM tempo
15:28 *🎤 Showcasing the AI's ability to generate dynamic vocals*
- Highlights its capacity for shouts, screams in the vocals
- Claims it has even better vocals than Udio
15:43 *🎙️ Sample of a rap song about AI/ML*
- Impressive rhyming and flow about topics like optimizing models
- Claims it passes the Turing test for being indistinguishable from human
17:34 *🔊 Demonstration of an AI-generated dubstep instrumental*
- Praises it as the best dubstep demo from an AI music generator
19:18 *🔮 Speculation about a new "text-to-song" AI tool*
- Founder claims to have tried an alpha tool that beats Suno and Udio
- Possibility of OpenAI having a revolutionary music generator like DALL-E
20:26 *📻 Overview of OpenAI's 4-year-old Jukebox music generator*
- Jukebox can generate songs from text prompts specifying genre/artist
- Despite being old, highlights the potential for modern improvements
Made with HARPA AI
thanks brother
Man, this is INCREDIBLE!!! That broke my heart song seriously gave me goosebumps. Gosh, thank you for sharing this!!!!
my pleasure!
I just published on my tiny channel a full Album, 10 consistent tracks, of a 70's rock band, all this created with Udio.
This new AI is stunning and thank you for bringing it to us :)
But Udio as a way more user-controlled approach to song writing.
This is mainly due to the fact that in Udio you create the song in 33 seconds chunks and, selecting the best one, extending it, tweaking the prompt for each extension at your will, you can build up a song with the structure you want.
Now Udio has new features of "inpainting", cutting, and so on and the control you have is incredible.
My personal opinion about generative AI, as a musician and digital painter, is that AI creative tools must give you the full control of the creation process. Creating a whole song from a single prompt is COOL, but apart from the prompt, the creator did very little.
For a comparison, Midjourney is by far the best image generator, but with Photoshop + Firefly you can create what you have in mind in a much more controlled way. Just my opinion, of course :)
Nice! Is your album out on Spotify/Apple Music? I'd love to hear it.
I've been playing around with Udio's new inpaint feature, but it's quite unstable. hopefully they'll update it to be more robust
Well said.
I am a video creator\editor and illustrator, so I agree about MJ. It's so hard to be specific with it, and I wish MJ had the same level of control PS have. it's inpainting is really limited and not enough tools to edit images, even if it's more for high level users. I know PS pretty well, so I can kinda work around MJ's limitations, but most users can't.
One major drawback is that you don't own the copyright to its creations.
Yeah but all someone needs to do is train another AI to use your unique techniques then you’re s__t outta luck 😅
@@AdrianMelia-0 You own the copyright to the lyrics if you wrote them.
Great video. Thank you for giving us all this information. I'd forgotten about jukebox and how good it was. Amazing what they were doing even 4 years ago!
you're welcome!
It is not dethroned until public gets to use it... we really don't know how much is assisted with those results. But with Udio go ahead and try something whack like oriental folk rap and the beast does it... and it's clean and works...
soon we wil be able to use this in Qbase for example and no need for online services :)
That part
This video is two month old and the AI is still not available to the public. Makes me wonder ...
I'm not sure how you train a music generation Ai without infringing on every major copyrighted song out there. I know there is or was a big lawsuit over Ai Music from the record labels.
Listen to my songs made with Udio
@@danieljones8706 listening to a song and learning from it isn't copyrighted imagine a human listening to a rock song then he goes and makes a rock song. The ai is only learning sounds and how they flow together in each genre it isn't sampling anything its literally creating new stuff.
@@DJ_Go_Hard Sadly, corps like money and they don't care about it (yet never pays their actual music creators worth over a fucking dime)
Remember last year when all the artists were selling off their catalogues for millions?
The Jazz one is all over the place- strange chord progressions, changing key haphazardly, shifting time signatures. Any real Jazz aficionado would laugh at the track’s discombobulation. The power pop song at the beginning, however was impressive.
Well any jazz aficionado laughs at any music so it’s not a good example lol
im not a jazz professional, so I cant review the jazz piece critically... but isn't jazz supposed to have "strange chord progressions, changing key haphazardly, shifting time signatures"?
@ai-tools-search yes jazz is supposed to be complex, but it also has a lot of structure that I don't think this particular example captured well. Even bebop, which is indecipherable to most casual listeners is generally based on a tune with a clear key center and a melody that works within that. This tune didn't really have that in my opinion.
The test for difficult music is to listen many times until you understand. Then you'll be able to judge whether the complexity is good or just random or wanky. I'm pretty sure this one will feel random to people even after fully understanding it, though bits and pieces sounded cool.
Bro, are you sure you're a jazz aficionado? Yeah, there were some unique choices, but y'know... that's jazz. There are guidelines but no hard and fast rules that mustn't be broken.
@@justinwescott8125 if the goal is to imitate big-band jazz it could be better; the piano harmonies don't match the brass and voice in the way that happens in 99% of this kind of jazz. If the goal is to make experimental music with prominent jazz elements, it was very accurate.
Well, that first 'pop' one "It started to sing" was pretty impressive. Better than most 'real music' that's coming out today. There were still a few little odd things in it and the lyrics could have been better. But the build up and instrumental sounds were quite good really.
Dude ur my #1 source of cool ai news
Thanks!
That widened tonality from the jazz song with the muddy hallucinating instrumentation gives it away for sure… 😂
That grunge song around 13:00 is awesome and it reminds me of a band of an old friend of mine Propeller. Really cool band!
Some time soon, when you walk into a restaurant, AI music will flow from the loudspeakers. Every song is listenable and fits the theme and the mood of the place perfectly, and you will never hear the same tune twice, because it is all generated on the spot.
you will just be able to ask your local AI waiter to make a few tracks that fits your night :) simple as that
You are correct.
@@nemesisone8927 And when it's time for the bill, the AI starts playing music inspired by the theme from Jaws :D
I really don't think so, at least in the near-term. Tech guys, by and large, underestimate how much people hate this.
@@augustuslxiii nobody hates this but musicians , and people that hate tech , also you wouldnt even notice the diference between manmade and ai , when its applied on ads and background music , nobody really cares
Dang bro, thanks for the update! Please make continuous contents about music ai 🙏🙏
my pleasure!
'Broek my heart' is very realistic -- I can't get that song out of my head, just like human-made pop music
My cellphone heard this and just became a groupy
I think I may like this 90s influenced song better than any song that actually came out of the 90s. It's cleaner grunge and mixed more to my tastes. I love how clean and upfront the drums and other instruments are in these songs. I hope we can access separate stems and vocals when we get into this so that we can make adjustments and COLLABORATE with the AI rather than just letting it do all the work.
Well that was obviously trained on a band called "Pavement" the voice tone and cadence are almost exactly the same, to the point where it sounds like a Pavement's song with chords shuffled around
Come on... they should rename jukebox to jukebot
Jukebot! Transform & roll out!
Hahaha. I wonder in 5 to 10 years some news outlet or journalist will use this buzzwords or meme. I just witnessed a meme in the making.
the 90s song just made me laugh ... because of how accurate it was...lol
Then E-books appeared, people said paper books will dissapear. I don't think music will die just like this.
Same when AI first beat a chess grandmaster
Same with the drum machine in the 80s.
not the music, but the musicians ...
In the hip hop space it can’t do very specific drums and Melodies according to a sub genre. N I tried suno and it keeps generating a different voice so it will be hard to capture and say “this is your artist”
That jazz jam was unreal 💜🎶🔥
indeed!
it was disgusting.
Everything blew my mind but the rap was the most impressive IMO. Completely indistinguishable from reality
Game of Thrones for music generation AI!
And wow! The thumbnail was NOT clickbait, +1 subscriber brah
The pop version is amazzing. The jazz version was clearly AI...ponderous and rambling, with offkey trumpet and overcooked chorusing. Broke my heart was quite interesting. Can these algorithms use supplied lyrics and or music to create the track?
What’s the purpose of music? It’s about evoking emotions. Can a song make you feel excited, melancholic, energetic, or sad? Can it make you stand up and dance? Can it make you reflect on something substantial? This should be the aim of music. I feel like many people wouldn’t care whether it’s AI generated or not. As long as it makes them feel something.
Another part of music, up until now, marks out events and contemporary culture in time. If the trend to AI shlock sticks, culture will stagnate
The battle is over. Humans had lost.
Haha, I love comments like this, remember, humans hade this!! xD we are actually winning af
Lmao I still love, listen to and support my favorite artists, visual & musical. These tools simply allow me to create more of what I like to hear. I'm sure many others feel the same. Digital art didn't kill traditional art, just like electronic music didn't kill acoustic music. There is simply more creativity to appreciate.
@@chillsoftjust because we made it doesn't mean we're winning, that's not how it works
@The1QwertySky it kind of is. I feel you guys exaggerate and are worried for nothing. Humans always win
@@ElsyOO366 i know AI cant do anything to us....yet. but AI is definetly better at almost everything than us
I am excited for this, I can't believe how good it is. I just got finished saying that about Udio.
It’s just the start. The meaning of this is that we are going to get divine music. abundance.
Ooo, it sounds very clear, nice voice sampling too.
„Turing Completeness“ has actually a different meaning in computer sciences. What you were thinking about is the „Turing Test“.
Thanks for this, awesome. I'm really looking forward to what's coming with AI music!
you're welcome!
It's interesting, you can hear the things it was trained on: Smells Like Teen Spirit, Have You Ever Seen the Rain, others... reimagined fairly seamlessly. But the inflection and timing is odd, and the structures/form/riffs are mundane.
I hear some Tom Petty in the first one as well.
I can hear one song or another, in every new human-made song in the last 12-15 years.
The 90's era singer is reminiscent of Conor Oberst (Bright Eyes).
@@OnigoroshiZero Well that's complete bullsh!t or maybe you just listen to trash.
@@mjames3662
Bullshit has no comparation end .
What no trash you listen ?
Good god this is amazing. The rap (Turing Completed) is crazy good.
I'm grateful I'm alive just to see this
great information. can't wait to try the elevenlabs music ai
I just finished a song with Suno that is pretty much radio ready. It won't be long before every ai platform will be making seamless stories, songs and videos no one will be able to differentiate with human works. I am a lyricist and I use my own lyrics with Suno, etc. providing the production. I love this stuff.
Lyricists will be honing their craft now
@@mr.guzwee7695 but it can generate its own lyrics
*GOOD LORD... THIS IS NOT EVEN THE BEGINNING OF THE JOURNEY AI WILL TAKE IN ALL FIELDS OF EVRYTHING... BEYOND TERRIFYING... THAT IS JUST INTENSE...WOW*
only a matter of time before AI can do anything better than us
actually, it's been quite a long ride ... I'd been training custom SRNN's since 2015 ... this is just commercialised bullshit spitting out quasi-randomised pastiches made out of of STOLEN uploaded work.
I wonder, at the end of the day what will be left for humans to do?
@@remor698play the video games all day long that Ai created for us.
@@MMoer Aka all we get to be is consumers... not exactly fulfilling for anyone who like me draws meaning in their life from doing something unique to themself (best way I found is sharing my creativity) that brings a positive light into the life of others and potentially inspires them to do the same. We each only have the one shot, best make it count towards something special.
It is amazing how far we have come with AI, but it will even get more interesting when AI is able to use your own voice, or when you can upload your own samples in order to let AI build a song around it, mixed over different tracks.
@@CharlieHufft-bk6noand you should own whatever the AI creates. You would also be able to spread your music allover the world with streaming apps
To think, just the other day, I had left some feedback stating that Udio sounded like something was... off. I'm... I'm not hearing that Uncanny Valley here. I guess this is the new way that life is going to be, now. And if I were to walk into some darkened cafe with this stuff playing in the background, I'd think that they have some indie band tunes for sale. What else can I say? Rest in Peace, Captain Dunsel of the music scene.
"M5 my beautiful creation, create a new musical opening theme that harms no one" M5 "Affirmative I will do as requested " Beep beeb.. "Engineering to the captain" Captain "Yes engineer this is the captain".. Engineer.."Captain Captain ! the AI beastie has taken over the ships studio and I can't stop it ! Science Officer here, "Engineer you can wallow in a pit of emotional despair fearing your obsolescence or remain logical and learn what is inevitable"
Sounds great, will try soon! 🎉
It takes a breath
🤯
@8:45 "Broke My Heart" 👀
That one was crazy! It's near commercial ready. it just needs a bit more work to get it there, but that's crazy this track was created from the ether by software.
I feel like only the slow people are worried about ai taking over artists, but for an artist like me, i see it as a tool that could help write or finish a song for you which is pretty sweet, also in the case of rap it is very subjective so you cant really replicate just one "rap style" like theres just far to many different genres of rap nowadays to be fully replicated yk
broke my heart to fill you up. i need this on my playlist
The future is generative
While we are degenerating
@@Happinessallmine soon we won't even be human anymore
Only way to save our job is being a specialist or some engineer and software engineer. Also where work with computer is just unfeasible due to high costs of the tech and internet connection needed.
@@cupofjoen There will be no jobs for humans
@@DG123zI’m fine with that☺️
Thanks for giving me a new perspective for those who are not good at mixing, I will learn to be able to produce my music with AI, hope you can give me some ideas so I can learn from you.
Broke my heart is for the top charts!
Elevenlabs knock it out of the park again big time. I was already keener on Udio of late, but a few minutes into watching this, I've paused and just cancelled my Suno plan. Looking forward to being able to upload audio guidance to generators, and midi generation too.
only tools in the wild please. if its closed then its not a proven threat to anything, just hype-ware until they can prove their game is real.
It’s from a company that already has a proven record of producing high quality AI voices. Add that ability to tech like Udio and Suno and you get the results shown.
Nothing here is surprising, in a few months there will almost certainly be significantly better.
@@lamsmiley1944 the problem with the TH-cam makers are they are so prone to click bait titles and thumbnails over substance. Udio and Suno are in no threat at all until this hype ware is proven real. Same for Sora, until it’s actually available. It truly does not matter.
@@elidelia2653 I agree that many creators (particularly those covering AI) use far too my hyperbole and click bait. I also think many people are overestimating the capacities of Sora.
I do however disagree that this is vapourware,
00:02 New AI music generator surpasses Udio & Suno
04:02 AI music generator creates full 3-minute songs from a single text prompt
07:57 AI music generator produces realistic jazz music with clean voice and instruments.
10:51 The AI music generator creates diverse music styles.
14:41 AI music generator produces realistic and dynamic vocals
16:58 AI music generator impresses with indistinguishable song quality
20:14 AI music generator is highly advanced and surpasses existing video generators.
23:05 OpenAI Jukebox can generate complete songs from text prompts and extend songs from short audio clips.
25:53 AI music generator's impact on Udio & Suno
Crafted by Merlin AI.
Got chills from the chorus ❤
Dubstep was over the top!
This music is actually good!
This directly competes with world-class professional musicians!!
All of it sucked. It's no wonder people like it.. Taylor Swift is popular, people like mediocre nonsense music Of course they will love this AI, it produces meaningless simple backround music like people already listen to
"Lines of Code!" 🤣
Wow that's pretty good not gonna lie! I def need to check it out.
I imagine a world in ten years where people will look at human made music and be like, "Ew, so much effort for that junk?"
In 10 years today will look prehistoric
You meant 6 months
Wow very very cool 11 labs have great voices and so making a making an Ai music generator will do great with other products they offer.
This is crazy! Now a full song on one prompt?? I was just getting used to thinking of myself as a clever song creator using Suno and udio, piecing together the various iterations into a beautiful cohesive whole, but now I’m not even needed for that!! Eleven labs can do the whole damn thing from beginning to end, intro, choruses, bridge, outtro and everything? It can create a masterpiece from a single statement! Our input as musicians are getting less and less. This is incredibly cool, and tremendously scary at the same time, depending on how you look at it.
that's what they claim. though these could be cherry-picked examples. i can't wait to try it out myself!
That just seems to be shortcutting using another ai platform to give you lyrics. To me this song started off strong and then got really generic.
Crap songs
ufffffff 😮the song MY LOVE is shockingly !!!!!! goood !!!!! im frightened!!!! this singer is exactly my taste … the drums sound great too. im really frightened
When do we expect them to release this?
Vocals are strong with elevelabs, but if you‘re focus is instrumental ambient or classic score music this kind of lacks a bit…. is the bass in the lows good enough?
Does it sound good in the clubs? 🤷🏻♂️
Great. This is a monstrosity. An afront to what it means to be human.
There are over 100 million long samples otherwise known as songs on TH-cam. With the ability to match the key, the tempo. and isolated vocals on the fly, you will not run out of new song combinations. Its litterally a new music business.
A sample is not a song.
@@Zoom1Zoom2-ix5kr how about a long sample
wave form for dubstep checks out
It sounded a lil bit like Tristan & Braken, bit of Skrillex idk. Find anything similar? 😅
awesome quality. i hope we can all try this soon.
I don't need it if it's not open source
💯
I didn't know elevenLabs did music. Do you have a link? I have reMade about 15 songs that I previously sung using Suno, I've posted most of them on this account, that last song I did on suno was almost 5 mins long. I can say your version sounds as good but why do you think it's better? I can see if people are using AI to write the lyrics, then it does save a step. But for people like me that have been writing songs for 20+ years, the end result sounds as good as suno, but not better. Song writers like myself need a lot more edit powers to customize the song generation, than these services offer. But I would gladly test it, can you provide a link? The only thing I find is Text to Speech type service.
It's not out for us to try yet. They've only released these demo songs and said that each was generated with a single prompt.
In terms of why I think this sounds better than Suno - the vocals are a lot clearer, and the overall track sounds cleaner. It seems that Suno generations have too much compression & saturation. Just a subtle difference. I'm sure there are a few Suno songs that sound great as well.
Not gonna lie, the song was absolutely 🔥🔥
This is really amazing
Jesus, what for? Could it file taxes instead? Do something we do not want to do, not taking our art and craft?
Great point ❤
Disagree
For real, what the heck ai can't file my taxes and do my homework from scratch before it take my job
@@tb620 What for? You fancy endless titillating stream of ai interpolated content?
People will be more able to express their ideas thanks to AI. Up until now making songs etc has been a lot of work.
Well, it's no surprise - it's from ElevenLabs, shout out towards Poland :D
Cmn, I thought Udio is the top, but there's more?! Damn!!!
haha I thought so too! it's crazy!
@@theAIsearch Lmao, can't wait to generate more orchestra music again in ElevenLabs!
Elevenlabs is still overpriced though…
Nice! Another tool to play with!
Jazz chords are way off (I play piano for 30 years), udio seem to make the chord progression better.
Suno seems the best for foreign language songs for now. Udio sings gibberish. Elevenlabs multilingual models should do an excellent job there too
can't wait for the renaissance of real musicians, songwriters and composers...thanks to AI it will help clean up the industry of an over saturation of so called "musicians" like beatmakers, computer desktop loop samplers etc. by making them basically obsolete.
you say thanks, but what will happen to those "musicians" then? what do they do if AI take over their job? AI won't replace the best of the best. But it will replace most of "musicians". That's billions of jobs lost.
@@SparkSphereIdeasAI is gonna give their power to the world
The truly uninspired and untalented beat makers who only use pre-made loops and pre-made kits probably will be replaced but I don't see it replacing ALL beatmakers. AI will definitely take market share though 100%.
The jazz band track was the only one that was obviously not right. The melody/chords were goofy and awkward and the instrumentation and lyrical continuity were rough, halting and not human - reminded me of a young composer learning what doesn’t work. All this makes sense tho since this is the most sophisticated style it attempted and it’s still learning. All that said, this is still stunningly better than anything I’ve heard to date and I have little doubt that I will be fooled in another year at most. Pretty crazy.
What a life to be a time
😃
Be a time to what a life
I use Eleven Labs for my professional voice cloning and that's incredibly realistic, insofar as AI goes, so I'm looking forward to checking out their Music generator, especially having used Udio, Suno and Aitubo in the past.
Now when can I generate songs this good on my PC?
i'm guessing there'll be an open source version in the coming months
Sanctum is already AI chat etc you can have on your pc :)
@@theAIsearch im sure it will come to DAW like qbase fruityloops etc etc they already have AI in them this will just be an addition :) cant wait just like photoshop has
I liked the vibratto from the jazz song, wow.
15:40 Eminem has been pretty quiet lately after this.
lol
I mean it's actually good give it 2 years and I guarantee you'll have your own pocket Eminem
he cant sing just like spears lmao all software to make it look like they can sing lol facts no fiction. "Autotune"
wow wtf broke my heart sounds soooo much like Tegan and Sara. Especially in that one part it even sounds like their voices. I don't know what to think of this. I mean i´d always rather listen to the real deal, but it would be good for making siamese dream like smashing pumpkins songs if you could train on those since Mr Corgan won´t do it 🙃
Soon AI will be composing magnum opus in the likes of Beethoven and great hits in any genre. Anyway, musicians won't be out of job yet because acoustic performances will keep having its place. It'll change when robots become common. In any instance, the audience, meaning most people, will benefit from it. Any musician must be excited. We'll have Elvis and Paganinis playing privately in our very homes. Extrapolating, our machine friends will deliver unique songs every time we need them. Later, we'll chip our brains so we can live musically all the time, i.e., in harmony.
THIS - is mind blowing!
what a way to destroy music industry even more and make music even less appreciated. i will continue to create music despite everything because I've dedicated all my life to it, and I don't care about these platforms creating generic music by stealing real humans work. just another nail in the coffin of society if you ask me. really want to unalive myself but i have too many responsibilities and it would be egoistical move. i hope these developers drown in the money that they are making by destroying lives of real musicians
If you actually want to unalive yourself for that reason alone, you should really talk to someone about this. That sounds like something way bigger than simply having competition in the form of AI.
Remember, people like when people do things. Therefore, AI music will likely only end up being used in the following three ways:
1. As generic background music in bars, when studying, etc.
2. As soundtracks for low budget projects (films and games etc) whose teams want to use their budget for other things than music.
3. As inspiration for people who work in the industry when they have writers block while on a tight deadline.
It won't take over everything. There is still room for human musicians and composers.
@@AnthonyBerlin i like your optimism, Friend, but when it comes to big industries, the worst scenario usually ends up being the way to go.
@@6oundStudio Well, for example, we still use orchestras when recording film scores even though most people couldnt tell the difference between that and virtual instruments that are much cheaper to use (at least if the virtual instrument version is done by someone very competent). Virtual instruments haven't destroyed the film music orchestras of the world.
Point is, these things doesn't always have to have the worst possible outcomes.
completely agree
do it
It’s really crazy what AI can do for music right now. Even in its early stages. As a musician I love it yet. I’m scared of it. I’ve put lyrics down and a couple of prompts and created some really awesome rock songs with udio and I’m thoroughly impressed, but there’s a part of me that wonders am I being creative or am I being lazy? Right now the way I see AI music generators are as idea starters. There a great way to take lyrics that you have no idea for a melody to, but you know maybe you want it to sound like ex-artist and you can get the ball rolling. And then if you are a musician, take those ideas and build upon them. I still hope there will be a human elements to music in the future. V 11 labs definitely sounds better, but all of these mixes have a bit of a artificial sound. Though they don’t sound bad enough that they wouldn’t make it on the radio. That’s just me saying maybe the stair didn’t crack how I wanted to or the kick drum didn’t white match the tone of the song, or the overheads sound artificial. Little things like that. Until AI can generate the nuances of reverb in a room, or the echoes of an auditorium, and the little acoustical ambience is better picked up in the Microphones I think most people will be able to perceive something is slightly off. people can hear an auto tune to voice. Even the best somehow sounds off. It’s great technology, but something deep down in the soul realizes it’s not totally human. It’s a strange direction we’re going, but there’s no stopping it. The only hope is that musicians will find a way to excel from AI. One day the human element will be the selling point at least for live performances.
That rap gave me a migraine.
I love This! Now more AI music. Been having a blast doing Suno Ai And Udio too. Now wow more of this yayy!
yes, more competition!