*Share your best results to my discord server!* ua-cam.com/users/redirect?event=video_description&redir_token=QUFFLUhqbE1oMFZ6RTlIUUVROFlqbkhfTmNldXdiMUhuQXxBQ3Jtc0trYnZYbEJBR0hkLUN0Z2hCRGE4cEg1NkdCZGtmdzdrUERHTHpNTi1nRWpGUVVkbm1hcmw2bWdlMDFocGdNQWMzR3hoWFBlNzJVdEd1SkZKMDNRVUtzUEIzM1AzempacVZERG1RWDZDM1gyZHlESFZpMA&q=https%3A%2F%2Fdiscord.gg%2FbQgcbjs2Sg&v=m39s703CitQ
ig that's what happens when someone does advertising for it x_x, servers die. But prob that's what you get for 4 bucks. The expensive ones cost more but better servers.@@mattskalicky
@chimpleton3484 bruh it's moving like lightning lol! This is full on generation, whereas that track (iirc) only had AI vocals that had a lot of training data.
Wait for the regulation.. All the music data AI use is protected by laws & copyrights, when new AI laws will kick in, those "tools" will be irrelevant, this will get regulated for sure
Yah, for a professional musician...it just meanders too much as a song. Perhaps sampling from it would be useful. I'd say the big time sample libraries should be crapping their pants about now...they'll be over in 1-2 years at this rate.
I don't think AI will be able to match the sound of sample libraries any time in many years from now. There's a huge logical problem with subtle instruments, as bowed string instruments for example. Even mathematics-based audio modeling libraries sound artificial, nowhere near the beauty of string instruments. What AI does is even worse, as it makes like an average of all, say, violins it has ever heard. But those instruments sound special because each of them has unique features. So at the end, to reproduce that realism, AI would have to be trained on the samples, the phrasings, the articulations of a single violin and then try to mimick it. In other words, first it seems necessary to have a perfectly sampled violin, which defeats the usefulness of AI. And even in this case, still for every note AI would take a dull average.
As a fellow musician, I agree about it meandering too much to really be useful. I tried a number of generations and each time, the songs would do weird things. I don't mean this in a bad way, but I'm not sure those behind the program are musicians or songwriters. The results I received lacked basic song structure. I'm impressed by its potential but feel its got a ways to go before it's truly useful for a musician. Right now, it's fun to play around with and to imagine the day when the big time libraries will be crapping their pants. Having watched this AI revolution from its early days, I'm continually astounded by the speed at which it moves.
@@davidmarshall8166 I agree with your observations. Actually, those who write these AI programs cannot teach anything to the AI. It's the AI itself that learns, by listening to millions of songs. That's the big limitation of AI: it's a black box, and nobody knows how it works internally, so one cannot modify it by hand. Notwithstanding million of dollars in training, for example, in image generations AI still has not clear that the human hand has five fingers... always, LOL. I think we are on a plateau now. It's a big lesson showing how much the cognitive system of the human being is important across all domain. It's not only pattern recognition, which AI can do, it's conceptualization which really matters. So I don't really expect AI to improve that much in the very next years.
AI music already sounds so good! It's gonna be awesome when you can just upload an existing song and have it automatically remixed in the style of your choice! Manually arranging even basic songs takes a really long time.
I love the stem Seperator. As a producer, I want to take an idea and make it my own and add around it with other instruments. If Suno AI did this, I would switch because of their quality of sound!
We need music upscaling. I feel like it could better handle hallucinating variation on existing track than creating one from scratch. If thats true, than I would think that a 2 step process of generating a decent sounding track and then upscaling it could fix issues like stitching. Maybe? We also need a "loopable" check box like some text ti image gen has "tilable"
Actually, now that I think about it, it would probably make those issues worse the way image upscaling creates bizarre stuff from the less defined parta of images.
Flute battle sounded cool !!! 😸 Cassette seem to have a very diverse music creation process that sets it a part somewhat from more mainstream generations. 👍
We need a controlnet in AI MUSIC like stable diffusion, where we upload a song, and Ai make variations of it adding new instrument, or remixing it with other style
It's interesting, though I wish it would let you upload midi as a source for the songs. I've been making a lot of them in preparation for the return of Musenet and you know...if it had that? If I saw that you could do that, that's an instant get for me. I do like that you can seemingly export as midi though. My project requires a specific chiptune flavor that I can't expect it to reliably produce and with a midi file I can just give it to it.
You should show how to do the editing/trim for this. I can't seem to figure it out. Any time I try to just trim a piece out the whole song just disappears from the edit section.
As a guitarist, and musician myself working on a song, I am impressed with this tool. I might even use it to make bass parts and synth parts for my song. It's really good!
Still doesn't hold a candle to suno, wonder what kind of secret sauce they're using. Also i tried generating two songs on cassette and 40 minutes later still generating lol
🎯 Key Takeaways for quick navigation: 00:00 🎶 *Introduction to Cassette AI* - Cassette AI introduced as a music generation tool. - Sunno AI comparison, discussion about updates, and expectations. 01:09 🚀 *Cassette AI Features Overview* - Overview of Cassette AI features and capabilities. - Sign-in process and initial exploration of the AI tool's interface. - Customization options, generation time, and song duration settings. 02:17 🎵 *First Song Creation: Beach Vibes* - Creation of a 2-minute song using the prompt "Beach Vibes." - Evaluation of the generated song's coherency and quality. - Discussion on AI weirdness and potential improvements. 04:50 🎵 *Evaluation of Another Song: Chill Loi Track* - Exploration of a 2-minute song called "Chill Loi Track." - Positive remarks on the relaxing and coherent nature of the song. - Recognition of slight stitching issues in the generated music. 06:34 💸 *Cassette AI Pro Plan Features* - Presentation of Cassette AI Pro Plan features and pricing. - Discussion on the affordability of the Pro Plan. - Decision to purchase the Pro Plan for further exploration. 08:52 🎧 *Analysis of a User-Created Song: 16-Bit Video Game Boss Music* - Creation and analysis of a user-generated 16-bit video game boss music. - Discussion on the impressive quality of the generated piece. - Introduction to the feature of separating individual stems. 11:33 🎵 *Separation of Music Stems* - Exploration of the ability to separate music stems (drums, piano, bass, etc.). - Excitement about the potential for personalized mixing. - Expression of surprise and interest in this unique feature. 14:53 🎼 *Attempt at Sound Effect Generation: Shotgun Reload* - Attempt at sound effect generation with the prompt "Shotgun Reload." - Acknowledgment of the model's limitations in sound effect generation. - Brief discussion on the capability to refine prompts. 16:30 🎷 *Exploration of Different Music Styles* - Exploration of various music styles: swing rhythm, jazz club atmosphere, EDM. - Evaluation of the generated tracks in different styles. - Recognition of the model's versatility in creating diverse music. 20:21 🌐 *Use of External Melody and Variations* - Integration of an external melody into Cassette AI for recreation. - Creation and evaluation of variations based on a successful generation. - Exploration of the tool's ability to handle multiple concurrent creations. 22:56 🎉 *Positive Summary and Recommendations* - Summary of Cassette AI features and capabilities. - Positive feedback on the AI tool's performance and versatility. - Encouragement for viewers to try Cassette AI and share results. Made with HARPA AI
Not too bad for instrumentals, though that chiming sound that starts around 3:55 sounds like what the CIA uses to interrogate prisoners. You'd probably get better results asking for 8-bit or lower or specifying a certain console. 16-bit is basically just regular music, Spotify streams its music with 16-bit encoding. Suno is still gonna be the king so long as it's the only one doing coherent vocals though.
It’s always funny how your expectations are so low that you like everything, even though it sounds like a confused raccoon playing the flute for the first time.
Matt. No offense but it's kinda weird how much you flip flop on music models. This is Google MusicLM level which we all agree is awful. Suno came out and you rightly called it better than MusicLM and now this is worse, unusable , incoherent and just not as clean as Suno and you say they should be afraid ? 🤯 I kinda get that you need to play the thumbnail/title game.
@@hipjoeroflmto4764 Meh, fair, I guess, but that's literally the difference between like... three keystrokes 🤷♂ I wouldn't call that "fear-worthy". Besides, the Suno devs have made it clear that with every update to the Chirp model, they plan on trying to increase the max length of each generation anyway.
Thanks for the heads up on all these new Ai tools. Sadly cassette is extreaaaamly sloooow for now, in fact im still waiting to finish generating something after 1 hour :-/ For now i'll stick to Suno
@@hipjoeroflmto4764 I don’t think so. Suno is really bad at instrumentals that it doesn’t shove vocals into in some way, especially video game instrumentals that sound anything like chip tune/8-16 bit. The AI doesn’t actually know any soundfonts nor was trained on VGM specifically, nor was it trained on specific styles outside of the dozen few that it has, so it’s pretty bad at video game music. If you want a unique “game” sound out of Suno, you’re probably not getting it. Cassette on the other hand actually has some potential, although it doesn’t actually sound “retro” at all, more akin to a Kingdom Hearts/Final Fantasy esque soundtrack, at least from that really good flute generation it had. If you want any legit VGM, your best route is to learn it yourself or pay someone else to make it for you, though that goes with any music for now, as AI hasn’t caught up to professional standards yet, let alone actual good, listenable music.
@@hipjoeroflmto4764 Excuse me? In no way was I being disrespectful, just giving my honest thoughts and adding to the conversation in an objective way. Not entirely sure how it came off that way. Also, why would that matter? I didn’t see your comment earlier than that because I haven’t seen this video since it came out, so I’m not entirely sure why replying later than you commented is bad in any way. If you’re somehow offended by a harmless comment about AI music, then that isn’t really my fault.
Much like AI art im quickly seeing that AI music has a kind of... smoothed over vibe. Like people are gonna know if music is AI generated. They may not be able to point out exactly why but people will quickly be able to point it out.
To my musically trained ears, the sound quality in Suno is infinitely better. I even managed to have it create the ambiance of a church for Baroque music. This is crazy.
@@HubertEtAlice I agree. I think the potential is there, but it's not quite ready for primetime yet. The results I received lacked basic song structure and meandered all over the place. I think it could be useful for sampling at the moment though.
@@davidmarshall8166 Sampling, yes, I thought the same thing. As for song structure, I think that Suno AI is very capable, given correct instructions. Her is an example I did (French Rock): ua-cam.com/video/NBpk2f1RKAA/v-deo.htmlsi=YqbMKYugNQ6VCPEL
The Estimated Time has run out and it's now saying "Almost done! We're under heavy load, please wait a few more minutes." The hell they are almost done. Why does MusicGen have to be non-commercial?!!
Hello, after watching your Suno video`s, i decided to go for a paid subscription. I`m not very good when it comes to music Genres! so i wonder if you? or someone else would help me? @8:12 you said it sounded like you was outside a club listening in. This reminded me of a sound i like! similar to what you said, but i don't know what type of music genera it is? best way for me to explain is... watch Madonna - Hung Up (MV) and the sound starts 25sec into the song. I love that type of sound, but have now idea what the name is called to describe that type of sound. Any help? anyone. Ty.
Meh. Compared to Suno Chirp, this is... okay-ish? Chirp has higher quality, no length limit (as long as you have enough credits for continuations), and of course the lyrics/vocals being the most important feature... On its own, this is nice enough, but in comparison, it's mid.
You don't need credits for continuations just disposable accounts and remix/continue ur tracks from ur other account there system is easily bypassable for free use
Hello, it’s me a new person. I would just like to ask, see if the end of the songs that are made in Cassett. at the end of them, do they feed out or do they have like a good ending or do they just cut off?
Is the song + music created with ChatGPT + SUNO AI private or it becomes a public property. Is there a way to keep it private on suno ? or does it becomes public property ? Thanks
Me too. I made 5 or 6 full songs so far, in different genres. I even made a music video (images done with Bing Image Creator, edited with Adobe Firefly, then video editing in Canva). I intend to make more of them. Piece of cake.
I checked it and not impressed at all. Just 7 songs per month! The quality of generated music and sfx too low. SFXs are just random noise for me. Music is very simple, not natural. Nothing of it can be used in any project.
Suno is obviously so much better. It kinda undermines your credibility that you cheese so hard on pretty much everything you review. And, Suno does indeed do instrumentals.
*Share your best results to my discord server!* ua-cam.com/users/redirect?event=video_description&redir_token=QUFFLUhqbE1oMFZ6RTlIUUVROFlqbkhfTmNldXdiMUhuQXxBQ3Jtc0trYnZYbEJBR0hkLUN0Z2hCRGE4cEg1NkdCZGtmdzdrUERHTHpNTi1nRWpGUVVkbm1hcmw2bWdlMDFocGdNQWMzR3hoWFBlNzJVdEd1SkZKMDNRVUtzUEIzM1AzempacVZERG1RWDZDM1gyZHlESFZpMA&q=https%3A%2F%2Fdiscord.gg%2FbQgcbjs2Sg&v=m39s703CitQ
Great job bro I’m loving it keep up the good work
Long a** redirect link
Audio separation AI has been pretty good for a few years now.
16:38 Close Encounters of the MattVidPro kind!
Bring back Uberduck or no
Suno is one of the only AI apps that still feel "magical" to me
Yesir.
Very powerful. And it can understand some meta commands, too.
Yeah...suno is the best at the moment...amazing!
Suno ai is like it was created not by humans but by extraterrestrial beings.
Right? Especially since you can so it right from your phone using your voice in seconds - ua-cam.com/video/wg6TyqmHn5A/v-deo.htmlsi=D7t_YCYPUVWjP623
please continue reviewing new and hot ai music models
Ofc!
Suno is still far better, and it can create good lyrics in any language
True, I was surprised it could sing in my native language of Chichewa of Malawi, a tiny country in southern part of Africa. 😊
@@apatsa_basiteniwhat does Africa smell like
@@hipjoeroflmto4764huh
It's hit and miss to be honest and it cuts off before the song finishes
I Wow, that’s amazing
Lot of it sounds like early 2000s cell phone ring tones 😂
Thank you so much for featuring CassetteAI Matt!
Your app is just MusicGen with an interface you are a scam
You guys are awesome
I've done 2 and I'm a pro-user and they both took several hours...
ig that's what happens when someone does advertising for it x_x, servers die. But prob that's what you get for 4 bucks. The expensive ones cost more but better servers.@@mattskalicky
I can't wait for music AI reaches the capabilities that stable diffusion does for images. I'm going to make some truly alien-sounding tracks
ikr its taking too long. we had drake ai a year ago
@chimpleton3484 bruh it's moving like lightning lol! This is full on generation, whereas that track (iirc) only had AI vocals that had a lot of training data.
Believe me, Suno AI can alrady do it. It is far more powerful than you think. I am sure the third version will blow everything off.
@@HubertEtAlice can I run it locally on my computer like Stable Diffusion?
Wait for the regulation.. All the music data AI use is protected by laws & copyrights, when new AI laws will kick in, those "tools" will be irrelevant, this will get regulated for sure
Yah, for a professional musician...it just meanders too much as a song. Perhaps sampling from it would be useful. I'd say the big time sample libraries should be crapping their pants about now...they'll be over in 1-2 years at this rate.
I don't think AI will be able to match the sound of sample libraries any time in many years from now. There's a huge logical problem with subtle instruments, as bowed string instruments for example. Even mathematics-based audio modeling libraries sound artificial, nowhere near the beauty of string instruments. What AI does is even worse, as it makes like an average of all, say, violins it has ever heard. But those instruments sound special because each of them has unique features. So at the end, to reproduce that realism, AI would have to be trained on the samples, the phrasings, the articulations of a single violin and then try to mimick it. In other words, first it seems necessary to have a perfectly sampled violin, which defeats the usefulness of AI. And even in this case, still for every note AI would take a dull average.
As a fellow musician, I agree about it meandering too much to really be useful. I tried a number of generations and each time, the songs would do weird things. I don't mean this in a bad way, but I'm not sure those behind the program are musicians or songwriters. The results I received lacked basic song structure. I'm impressed by its potential but feel its got a ways to go before it's truly useful for a musician. Right now, it's fun to play around with and to imagine the day when the big time libraries will be crapping their pants. Having watched this AI revolution from its early days, I'm continually astounded by the speed at which it moves.
@@davidmarshall8166 I agree with your observations. Actually, those who write these AI programs cannot teach anything to the AI. It's the AI itself that learns, by listening to millions of songs. That's the big limitation of AI: it's a black box, and nobody knows how it works internally, so one cannot modify it by hand. Notwithstanding million of dollars in training, for example, in image generations AI still has not clear that the human hand has five fingers... always, LOL. I think we are on a plateau now. It's a big lesson showing how much the cognitive system of the human being is important across all domain. It's not only pattern recognition, which AI can do, it's conceptualization which really matters. So I don't really expect AI to improve that much in the very next years.
Cassette AI will be the game changer for the industry, at this early stage it is way ahead than others
AI music already sounds so good! It's gonna be awesome when you can just upload an existing song and have it automatically remixed in the style of your choice! Manually arranging even basic songs takes a really long time.
Yo Matt, had to wait for a couple minutes.. but the creations were freakin awesome. Love the video
I'm kind of surprised Suno hasn't gotten like mainstream attention yet.
I love the stem Seperator. As a producer, I want to take an idea and make it my own and add around it with other instruments. If Suno AI did this, I would switch because of their quality of sound!
Waiting those few seconds is really worth if you are getting good music the way you want
We need music upscaling. I feel like it could better handle hallucinating variation on existing track than creating one from scratch. If thats true, than I would think that a 2 step process of generating a decent sounding track and then upscaling it could fix issues like stitching. Maybe?
We also need a "loopable" check box like some text ti image gen has "tilable"
Actually, now that I think about it, it would probably make those issues worse the way image upscaling creates bizarre stuff from the less defined parta of images.
Mario?
@@hipjoeroflmto4764 ? Huh?
@@Agispsi waaaah
@@hipjoeroflmto4764 cool story
Coproducer output AI is a giant sample generator, creates a mix with 10+ samples based on text, and gives you all the files (30 second samples)
Suno is definitely still king, mind blowing in ways. But the track separation in cassette is a great feature.
Flute battle sounded cool !!! 😸 Cassette seem to have a very diverse music creation process that sets it a part somewhat from more mainstream generations. 👍
We need a controlnet in AI MUSIC like stable diffusion, where we upload a song, and Ai make variations of it adding new instrument, or remixing it with other style
15:37 deaf after a loud sound or horror scene or aliens before your vision fades out to white.
perhaps Suno and Casette could team up (Suno for the vocals and lyric generation and Casette fir the instrumentals)
Estimated Time: 1738 Seconds. Wow...they getting slammed! will be fun to play with if they can scale it effectively.
Really stunned by track seperation feature of cassette, it is way ahead than others.
It's interesting, though I wish it would let you upload midi as a source for the songs. I've been making a lot of them in preparation for the return of Musenet and you know...if it had that? If I saw that you could do that, that's an instant get for me. I do like that you can seemingly export as midi though. My project requires a specific chiptune flavor that I can't expect it to reliably produce and with a midi file I can just give it to it.
You should show how to do the editing/trim for this. I can't seem to figure it out. Any time I try to just trim a piece out the whole song just disappears from the edit section.
As a guitarist, and musician myself working on a song, I am impressed with this tool. I might even use it to make bass parts and synth parts for my song. It's really good!
Still doesn't hold a candle to suno, wonder what kind of secret sauce they're using. Also i tried generating two songs on cassette and 40 minutes later still generating lol
Aliens are real bro.
But there corporate
@@hipjoeroflmto4764 Same
🎯 Key Takeaways for quick navigation:
00:00 🎶 *Introduction to Cassette AI*
- Cassette AI introduced as a music generation tool.
- Sunno AI comparison, discussion about updates, and expectations.
01:09 🚀 *Cassette AI Features Overview*
- Overview of Cassette AI features and capabilities.
- Sign-in process and initial exploration of the AI tool's interface.
- Customization options, generation time, and song duration settings.
02:17 🎵 *First Song Creation: Beach Vibes*
- Creation of a 2-minute song using the prompt "Beach Vibes."
- Evaluation of the generated song's coherency and quality.
- Discussion on AI weirdness and potential improvements.
04:50 🎵 *Evaluation of Another Song: Chill Loi Track*
- Exploration of a 2-minute song called "Chill Loi Track."
- Positive remarks on the relaxing and coherent nature of the song.
- Recognition of slight stitching issues in the generated music.
06:34 💸 *Cassette AI Pro Plan Features*
- Presentation of Cassette AI Pro Plan features and pricing.
- Discussion on the affordability of the Pro Plan.
- Decision to purchase the Pro Plan for further exploration.
08:52 🎧 *Analysis of a User-Created Song: 16-Bit Video Game Boss Music*
- Creation and analysis of a user-generated 16-bit video game boss music.
- Discussion on the impressive quality of the generated piece.
- Introduction to the feature of separating individual stems.
11:33 🎵 *Separation of Music Stems*
- Exploration of the ability to separate music stems (drums, piano, bass, etc.).
- Excitement about the potential for personalized mixing.
- Expression of surprise and interest in this unique feature.
14:53 🎼 *Attempt at Sound Effect Generation: Shotgun Reload*
- Attempt at sound effect generation with the prompt "Shotgun Reload."
- Acknowledgment of the model's limitations in sound effect generation.
- Brief discussion on the capability to refine prompts.
16:30 🎷 *Exploration of Different Music Styles*
- Exploration of various music styles: swing rhythm, jazz club atmosphere, EDM.
- Evaluation of the generated tracks in different styles.
- Recognition of the model's versatility in creating diverse music.
20:21 🌐 *Use of External Melody and Variations*
- Integration of an external melody into Cassette AI for recreation.
- Creation and evaluation of variations based on a successful generation.
- Exploration of the tool's ability to handle multiple concurrent creations.
22:56 🎉 *Positive Summary and Recommendations*
- Summary of Cassette AI features and capabilities.
- Positive feedback on the AI tool's performance and versatility.
- Encouragement for viewers to try Cassette AI and share results.
Made with HARPA AI
You should have tried the person falling down the stairs while eating an apple
Cassette is good with the instruments and has potential to produce good music
Not too bad for instrumentals, though that chiming sound that starts around 3:55 sounds like what the CIA uses to interrogate prisoners. You'd probably get better results asking for 8-bit or lower or specifying a certain console. 16-bit is basically just regular music, Spotify streams its music with 16-bit encoding. Suno is still gonna be the king so long as it's the only one doing coherent vocals though.
Played around w it, video prompting is unreal. Feels more in depth than anything else I've tried
It’s always funny how your expectations are so low that you like everything, even though it sounds like a confused raccoon playing the flute for the first time.
Nowhere close to Suno, but for sampling definitely perfect. Tried the model it's not working, it's stuck at Almost done while loading the audio.
Matt. No offense but it's kinda weird how much you flip flop on music models. This is Google MusicLM level which we all agree is awful. Suno came out and you rightly called it better than MusicLM and now this is worse, unusable , incoherent and just not as clean as Suno and you say they should be afraid ? 🤯 I kinda get that you need to play the thumbnail/title game.
They should be afraid it can generate longer audio than 1:20
@@hipjoeroflmto4764 So can Suno Chirp. You can do as many continuations as you want, to get as long a song as you want.
@@IceMetalPunkwithout reprompting.
@@IceMetalPunktell me you are an asshole without telling me you are an asshole
@@hipjoeroflmto4764 Meh, fair, I guess, but that's literally the difference between like... three keystrokes 🤷♂ I wouldn't call that "fear-worthy". Besides, the Suno devs have made it clear that with every update to the Chirp model, they plan on trying to increase the max length of each generation anyway.
What ai tools can add voice clonde lyrics?
crazy stuff!!! great content man🤘🏻
15:51 beautiful
Is it meant to take an hour to generate one song? Because that is what it is doing to me. Haven't done anything yet
Thanks for the heads up on all these new Ai tools.
Sadly cassette is extreaaaamly sloooow for now, in fact im still waiting to finish generating something after 1 hour :-/
For now i'll stick to Suno
Tried CasetteAI, Suno, have experience w stable audio. Cassette def made more towards musicians I’d say
Seems like it's best at gaming soundtracks! This is a goldmine for game developers!
Suno is better for that, u can get instrumentals from it
@@hipjoeroflmto4764 I don’t think so. Suno is really bad at instrumentals that it doesn’t shove vocals into in some way, especially video game instrumentals that sound anything like chip tune/8-16 bit. The AI doesn’t actually know any soundfonts nor was trained on VGM specifically, nor was it trained on specific styles outside of the dozen few that it has, so it’s pretty bad at video game music. If you want a unique “game” sound out of Suno, you’re probably not getting it. Cassette on the other hand actually has some potential, although it doesn’t actually sound “retro” at all, more akin to a Kingdom Hearts/Final Fantasy esque soundtrack, at least from that really good flute generation it had. If you want any legit VGM, your best route is to learn it yourself or pay someone else to make it for you, though that goes with any music for now, as AI hasn’t caught up to professional standards yet, let alone actual good, listenable music.
@@cloudirubez07 no I'm not giving a thoughtful response to somone who wants to reply to me after 4 weeks bye
@@hipjoeroflmto4764 Excuse me? In no way was I being disrespectful, just giving my honest thoughts and adding to the conversation in an objective way. Not entirely sure how it came off that way.
Also, why would that matter? I didn’t see your comment earlier than that because I haven’t seen this video since it came out, so I’m not entirely sure why replying later than you commented is bad in any way. If you’re somehow offended by a harmless comment about AI music, then that isn’t really my fault.
@@cloudirubez07 okay well I have no part of it anymore it's old for me. Have a good day or night whatever part of the world ur in.
Much like AI art im quickly seeing that AI music has a kind of... smoothed over vibe.
Like people are gonna know if music is AI generated. They may not be able to point out exactly why but people will quickly be able to point it out.
10:15 Wow! This is so awesome! Even in 2x speed!
"Estimated Time: 3609 Seconds"
Oh, great! Look what you've done, Matt! XD
I like the "droning out" part.
Doesn't work unless you pay for upgrade. Why would I pay before even trying it? Looks like Suno is the best option for now
I have instrumentals and Im trying to put Lyrics and Vocals on them
... Anybody know a ai for that ??
After having hands on experience, I could say music is accurate to the prompt given just it takes bit time.
What's the best for movie score?
LOL the sites overloaded :)
Music produced by Cassette is far better than others
To my musically trained ears, the sound quality in Suno is infinitely better. I even managed to have it create the ambiance of a church for Baroque music. This is crazy.
@@HubertEtAlice I agree. I think the potential is there, but it's not quite ready for primetime yet. The results I received lacked basic song structure and meandered all over the place. I think it could be useful for sampling at the moment though.
@@davidmarshall8166 Sampling, yes, I thought the same thing. As for song structure, I think that Suno AI is very capable, given correct instructions. Her is an example I did (French Rock): ua-cam.com/video/NBpk2f1RKAA/v-deo.htmlsi=YqbMKYugNQ6VCPEL
This was horrible. Didn't deserve any attention to be honest. Suno is ages ahead.
Matt when are you going to make a video about "Ai text to food" im getting hungry
Ewww
Same
I have tried both suno and cassette and i can say cassette gives more freedom to create something great.
Excellent Reaction!
Cassettte>Suno. More control over what you create
You can also make instrumental music with suno
Indeed you can. Just put [No lyrics] where you are supposed to write the lyrics.
Your so enthusiastic about mediocre stuff. Must be amazing to be moved that easily. I find I need a lot more to get excited.
The Estimated Time has run out and it's now saying "Almost done! We're under heavy load, please wait a few more minutes."
The hell they are almost done.
Why does MusicGen have to be non-commercial?!!
I'm glad you liked it but honestly I thought it was pretty bad. 😂
Uhm. That's not music. Just a pile of sounds with a hint of harmony.
Nah, that's not even close to Suno.
can it make chiptune?
Hello, after watching your Suno video`s, i decided to go for a paid subscription. I`m not very good when it comes to music Genres! so i wonder if you? or someone else would help me? @8:12 you said it sounded like you was outside a club listening in. This reminded me of a sound i like! similar to what you said, but i don't know what type of music genera it is? best way for me to explain is... watch Madonna - Hung Up (MV) and the sound starts 25sec into the song. I love that type of sound, but have now idea what the name is called to describe that type of sound. Any help? anyone. Ty.
I would say probably French House, or Nu Disco. That low pass filter sound reminds me late 90s Daft Punk.
@@jackied962 Hi... Thank you kindly for taking the time to reply back to me. I think it`s more Nu Disco. Thanks.
Just waiting for one that makes VGM.
I literally generated 3 songs with musicgen and this cassette Ai didn't generate anything. It's stuck on "Almost Done"
I think we melted the servers
Meh. Compared to Suno Chirp, this is... okay-ish? Chirp has higher quality, no length limit (as long as you have enough credits for continuations), and of course the lyrics/vocals being the most important feature... On its own, this is nice enough, but in comparison, it's mid.
You don't need credits for continuations just disposable accounts and remix/continue ur tracks from ur other account there system is easily bypassable for free use
Hello, it’s me a new person. I would just like to ask, see if the end of the songs that are made in Cassett. at the end of them, do they feed out or do they have like a good ending or do they just cut off?
The generation time is painfully slow
i think suno is better, cassette to much strange sounds
Is the song + music created with ChatGPT + SUNO AI private or it becomes a public property. Is there a way to keep it private on suno ? or does it becomes public property ? Thanks
Yourtracks remain private anyway, but if you want to comercially use your tracks, then you have to go for a subscription plan.
@@HubertEtAlice Thanks
It's broken. Probably getting smashed.
Will ai replicate cgi people from movies
21:34 the facial expression hahahaha
It took like half an hour to generate a single song and it sounded like shit lmao.
I keep clicking on these...but i have not heard anything good yet
oooo new music ai :0
Yee
cool, thank you!
Teach the system to output actual **Stereo** musical-soundtracks. I'll plop down the buckos for a Sub, then.
Crazy
Suno still better, Im making a whole song thanks to Suno, and I mean a good song!
Me too. I made 5 or 6 full songs so far, in different genres. I even made a music video (images done with Bing Image Creator, edited with Adobe Firefly, then video editing in Canva). I intend to make more of them. Piece of cake.
Sorry, but I don't like this at all. I think it is annoying. Suno is waaaaaaay better.
15:56 21:39 22:15 MY EARS!!!! XD
Elevator music generators
Wow, that is absolutely horrendous. Doesn’t stand a chance.
Can you download each element and then use the mixer ?
I mean, recording each element works for sure.
I would think if you have an account you can download each instrument maybe @@metalmonkey128
The music is all over the place. No structure, melodies are oddly random. This is pretty awful sounding.
Suno is way better than this, and it creates song lyrics as well as lets you attempt to write your own
I love cows
16h 677 likes 239k shsb
I think suno was definetly better.
As much as i think Suno is good, this one is average.
It can't do songs
Sunos way better tho right?
Yes
I checked it and not impressed at all. Just 7 songs per month! The quality of generated music and sfx too low. SFXs are just random noise for me. Music is very simple, not natural. Nothing of it can be used in any project.
Trash Beach Vibes 😅🗑️ 😷🤢 Trash 💯
Beaches have trash
Yeah.... Nah bro. It does not, in any measure, go hard 😂
Nice
Suno is obviously so much better. It kinda undermines your credibility that you cheese so hard on pretty much everything you review. And, Suno does indeed do instrumentals.