Congrats for your channel. Useful, well produced, well edited, fun and informative. I am an intermediate guitar player myself. Although I can play most of the parts of different instruments, I have a hard time putting them all together in sync. Your videos already show a lot but could you please do a “live”, real time creation (no edits) of a song so we can watch your step-by-step full workflow?
Thanks for watching and I'm glad you're enjoying the content! Teddy has been a fantastic sleeper. We got really lucky. He goes to bed every night around 9pm and sleeps all the way until 7am. I know that might change but I hope not!
This is by far the best video that I've seen that demonstrates what this AI tool can do. Like others, I think it's really good sounding as a backing vocal but not too sure about lead vocals. Also, it seems geared for pop. Do you know of any real implementations of rock vocals?
Thank you so much! I recently did another review of a service called "Audimee". It's a converter, so you'll need to record you own voice as a placeholder and then convert it to one of their rock voices. Audimee's rock vocals are super good, 100% recommend!
My opinion? I think it's a good way to demo songs, especially vocals, if the singer is having trouble, we can create midi vocals, then the singer has something to sing with while recording. Also if we create songs, we can show the ideas to our band that we have on mind. But something is missing, when the midi vocal sings the high parts, it doesn't sound right, sounds fake, that's why the real vocal will always show the emotions better.
They often call that a scratch track. I agree, this is really great for that. An actual recording still feels and sounds better, but this is so great to convey an idea or example!
I agree that the emotion of a real performance is something that AI has yet to fully capture. It is a great tool to plot out ideas though. It's very empowering!
as for your question i think a blend between your voice and the AI's is the best way to go. I believe AI in the future will be utilized best that way. A hybrid basically, and if you can't sing. Well then perhaps work really hard on the AI to make it absolutely spot on! I love the vibrato tool it looks amazing, can you bend up the notes too? If not i'd suggest you tell the developers that since that is a common human way of singing.
Yes you can manipulate the pitch any way that you see fit. There's a ton of control inside the program- it can actually be a bit overwhelming at first. Once you get the hang of it it's pretty straightforward though.
This is both shocking and kind of fantastic. It's actually the UI/UX which has me most excited. Good voice cloning has been a thing for a while now, but it has never been so usable. What I would love to see is how this pairs up with one of those sing (or hum) to MIDI apps to generate notes, and then take that MIDI and send it into ACE. That would at least theoretically allow someone who can't sing to have something in their head and with some effort get to hear what it actually sounds like when it isn't being murdered.
The UI has come a long way and the level of customization is absolutely wild. I love that this tech is empowering people to make the music they hear in their heads- it's absolutely awesome!
Spencer, I love that you constantly experiment with new technologies to stay ahead in our ever-changing world. Nothing matters more than relevancy. I appreciate your talent, passion, creativity, and refusal to relinquish your dream despite adversity. It is beautiful to watch and listen to! Moreover, your vocals sound better than the AI vocals because I can instantly detect their robotic sounds. I have always enjoyed your singing voice and unique delivery. Although this software would be useful for a songwriter pitching a quality demo, especially in a female vocal dynamic, if the song calls for that and the user is a male vocalist or has no singing ability. Combining your human vocals in Logic with the perfectly-pitched AI tracks panned opposite each other would make the finished song sound more realistic and better. Thank you for introducing me to this new technology. I look forward to your thoughts on my insights! This is my favorite RUclips channel, and I can see that your fans and followers are growing by the minute. Please don't ever stop doing what you do. You are helping so many people!
Your support means so much. Making these videos is a ton of work but I absolutely love making them, and comments like these are why I do it. I firmly believe that anybody can make music, and the fact that people are taking something away from my videos is incredibly rewarding. So thank you for sharing that, it made my day! I do agree that the roboticness can be noticeable at times. I think if I spent a bit more time fiddling with it, it would be better. I agree that a hybrid approach is probably the way to go. I think recording a real vocal and using the AI for harmonies would give us the best result!
Wow. Very impressive! For your banger, I liked your performance better because it projected more energy and intensity at the high notes. The AI singer felt like holding back. This can probably be fixed with more editing in ACE studio. But both performances rocked!
So glad you showed the consonant sliders and other tweaking. Many videos make it look like you just type in the lyrics and you're good to go. You can spend a lot of time fiddling with the words. Deep voices are not as good and long sustained notes can be very iffy. That said, I'm having some good success making choir rehearsal tracks....would like some voices that are less pop or hip hop for choir.
Thanks! I agree, I'd like to see more versatile voices. Rock voices would be cool, as well as more classical style voices. I'm sure it's only a matter of time before they're available!
My answer to your question: of course your voice, Spencer!! Big soul, excellent talent, georgeous videos!! Thank you very much for all your hard work. My best wishes for you and your beloved ones, from Spain ✨🤗✨
Works absolutely great with your voice but then you're a very good singer 🙂. In this case I'd choose your lead combined with ACE harmonies, that opens up interesting perspectives regarding formants. In general I'd go a little easier on the vibrato in ACE, if you start doing it on every long note it might fast become unnatural. IMHO that is 🙂.
It is a powerful and very useful tool, but I do prefer your voice a million times over the AI examples. I would use it more as a demo creation tool for giving the melody guide to a singer for a project to let them know what I want for the song. The timbre, genuine feel, and overall performance could never be replaced by any AI no matter how good. Just like you said, this is just my humble opinion. By the way what a charming and lovely tune the one one you wrote for your son. Blessings for you and your family and thank you for such amazing content always!
Thanks for sharing your insights and the kind words! it means so much. I do agree at the very least, these AI tools are fantastic for developing ideas and creating demos. I love being able to whip up a woman's voice right from my computer, it's incredibly convenient and creatively empowering.
Are you still subscribed to ACE Studio? Try raising the Air parameter for the reduced airflow to make Naples super powerful, there's the rock/metal potential
As far as I know it's only available as a separate application currently. I'm having trouble finding the system requirements unfortunately. If you have a decent computer (16gb of ram) and it isn't super old I would imagine it would run just fine.
@ I’ve got a answer 1. Compatibility with iMac M1: Yes, an iMac with an M1 chip is more than sufficient to run ACE Studio smoothly. The M1 chip provides excellent performance and efficiency, so you should have a seamless experience while using the software. 2. Storage Requirements: To ensure optimal performance, we recommend having at least 5 GB of free disk space for installation and smooth operation of ACE Studio. 3. System Requirements: Operating System: macOS 11 or later Processor: Apple M1 chip (or Intel-based processor with a minimum of 2 GHz) RAM: At least 8 GB (16 GB or more is recommended for better performance) Disk Space: 5 GB of available space for installation and operation DAW Compatibility: Works with most major DAWs including Logic Pro, Ableton Live, and others
I think I would love to use this if I'm strapped for time and want to add a lot of harmonies. This would prevent me from doing so many takes, BUT I would want it to be my voice (in most situations, although I am not opposed to other voices as well). Is there a voice upload/record option? For lead vocal, I would want it to be organically me. So I'll sing that one myself :)
I prefer your voice because on the "Woah's" there was just a little bit of grit as you leaned into it. The Spencer voice had a touch of it, but not enough in my opinion. Is there a way to dial that up with the AI voices? BTW, loved the track itself!
Spencer, your voice is clearly better!! I don't know the technical terms to use, but it sounds like your high notes are more "shiny". Perhaps it's the human placing the voice slightly differently compared to the AI voice.
@@johnnyfilippone Spencer didn't really properly used the energy, tension and note bends on the ai to simulate that high end so its not a fair contest. I been using it and you can't just spam notes. You gotta start from a down note and take advantage of its capabilities to realistically pitch bends and energetically raise at certain parts of the singing.
Naples doesn't really have the "shiny" bright tone to start with, although that makes him in a weird position for where he's still best suited for the 4th octave source: am an ace studio user
This sounds great. I have tried to find some Ace Studio examples of rock vocals to use in my productions. Does Ace Studios have rock vocal examples, and if so, can you point me in the right direction?
I haven't had the best luck creating Rock Vocals with Ace Studio. Most of their voices are best suited for Pop. I'd like to see some Rock Voices added. Hopefully they do so in the future!
Yamaha's VOCALOID has been a legend in this field since 2003. The difference is that back then, the AI tag was not used, whereas nowadays, the AI tag is prominently associated with it. VOCALOID 6 can also be used as a VST plugin in your DAW.
I used Vocaloid a few times as examples for real singers to have an idea of what I was aiming for and for that it worked wonders but Vocaloid is "miles" away of this new AI stuff.
I still hear some areas that aren't quite convincing. But this software seems to come with enough tools to tweak it until it does sound more realistic. It's certainly already leaps and bounds above Emvoice, which I've been using a bit for backing vocals. That being said, I suspect we're really not that far away from truly convincing AI vocal tools. Not sure if that's a good thing or not.
I agree that there are some parts that could be more convincing, but like you said- the tools are mostly there. It's going to be absolutely wild to see where this technology is 5-10 years from now. It's super scary yet exciting
@@songsbyspencer Yeah. It kinda reminds me of how, 25 years ago, I was convinced that there would never be a guitar emulation that would sound authentic. Which sucked because I can only play keys. About 5 years ago, I decided to Google it and see if there was anything half decent. I was blown away! Now I've got several guitar/bass/drum VSTs, and I started working on a bunch of ideas that sat on the back burner for years.
You can create your own voices by uploading data and modeling it- so that's pretty sweet! As of now they support English, Spanish, Chinese, and Japanese- so no Portuguese just yet. I'm sure it's only a matter of time.
@@songsbyspencerMuito bacana da sua parte separar um tempo para responder a todas as perguntas, ainda mais levando em conta que você tem uma criança pequena para cuidar.
This needs to be available as a plugin for DAWs, rather than a stand alone app that requires all that importing/exporting. That's not the optimal workflow if you're creating and experimenting
It does work inside a DAW via a VST plugin. It's called ace Bridge I believe and it is located in the Ace folder. Just copy the dll and paste it in the vst folder that your DAW is pointed to for vsts. I use Reaper and this worked for me.
AI can't even come close to having a nice "growl" like yours, Spenz 12:01 😁... but this ACE Studio would sure give Dreamtonics Synth V a run for their money 😆 great demo from you btw
Amazing program bro! Can I turn this program an 80's metal singer? I already use A.I. to make music for the Bible. Later (I hope) when I will have more free time I would make tons of songs in 80's metal styles.
If you export the audio of your track, and pull that into Ace Studio, you could listen to the music behind the vocal being generated, and edit as needed.
It’s a step in the right (or wrong) direction. But I think the voices available are very mediocre at best. But if they got decent vocalists, then it could be a game changer if it works as good as you’re portraying in this video.
AI voice changed based on your recording would work so much better as you have some really nice inflections and tonal slides that is not captured in the MIDI, making it sound less natural and realistic. Great demo though!
In the video I plug at the end of this one, I review Audimee- which does exactly that! It transforms your vocal takes into different voices, but maintains the same characteristics. It's awesome!
The female voices on this platform seem stronger than the male. The other AI site you reviewed had great male vocals. For the “banger” I liked your vocals better, but want this site does really well is the vocal harmonies….when the voices are layered they sound even better…
I agree! As with all AI tools I think they're best used when they complement the real thing. This has been super useful when layered in the background!
@@songsbyspencer Yes, I tried it out. It works. Not with musicxml, but with MIDI. Lyrics get exported in MIDI. I didnt know that up to now. Thanks for the reply!
I would totally use this for demos. I would not production release this with AI singing lead or even backing vocals because the service that does this can potentially file copyright claims for the use of their product. Don’t like the legal implications for this for anything other than personal or demo use. Also, who’s going to buy an album knowing that it was AI-generated? That cheapens the value of YOUR music. Demo? Yes. Production release? Hard no.
Very fair point. I really hope AI is used as a creative tool as opposed to making artists obsolete. I could see it going either way. You raise a great point though, I don't think people will be interested in listening to AI music. As a novelty, sure. But listening to an hour long album that's entirely AI would put a lot of people off.
I appreciate that! I"m glad you're using it and I hope it's been useful for you. I really like being able to generate female vocals, that's my favorite part. Emma is my favorite voice!
I love your videos and I dig the channel, but man... to me this AI stuff is a bummer. I get it; the tools are out there and I guess someone has to showcase them, but it doesn't do anything for me, at all. Also, you just wrote lyrics and melody but uploaded it to the machine for it to sing: Who owns the idea now? Genuine question. Anyway, happy holidays and all the best.
I can totally see where you're coming from. This technology is absolutely incredible, but it's also scary. I see the angle of how AI might devalue a real producer's work, but I also see the angle of how it lowers the barrier of entry and gets more people to start making music. I'm optimistic and I'm not overly concerned with AI taking over music production. Creating great music requires a lot of creativity, and it's an extremely nuanced process. If someone were to only use AI tools and premade loops to create their song, it isn't going to stand up against a producer who creates each element of their song by hand. A real producer feels an instinctual need to make music, and they will do so with whatever tools they have available. That passion always shines through, and it's something I have yet to see AI come close to replicating on a large scale. It can make pretty convincing vocals if you tell it what to say, what melody to follow, and you work with it. But for it to come up with all of that on it's own? I have yet to see it. But who knows, 5 years from now I might eat those words! Also, the creator of the lyrics and melody is the writer of the song by definition, so I don't think there are any legal issues there. I know most of the voices in Ace Studio don't require a license and are royalty-free. There are a couple voices that require additional licenses to use, but they're clearly marked to avoid confusion.
Atm monthly subscription is U.S.$22/month, and the annual one is U.S.$149/year. That will stop the hobbyist, but so it was with Ai-image creators, so .. A 'week' from now, and it free.. The 'battle' Imo a draw, but exact at the beginning Spensers voice had more ..i dont know *appeal* i guess it could be called
This is how AI could be used for composers that can't sing at all. I wish i had this as a kid, imagine the kids growing up with this tech. Please kids, make amazing music again. Now the dreams the possibilities are endless, the only thing threatening your amazing songs is the record labels, but hopefully they'll lose against the AI companies. AI is doing imitations just as we humans do. When i play guitar i try to imitate Gary Moore all the time should i be brought to court now? Interesting times to be alive!
Spencer you didnt really use the voice editor to the max. It can pitch bend and slide up notes but you did it word for word and straight note key for key which made it sound ai. Your vocals you were using pitch bends and all, you were pitch shifting and doing all the emotions which you didnt edit into the ai. It shouldnt be completely 100% note for note is what i am saying on the ai edit. Theres a lot of different notes in 1 note. Ill send you an example when i have a chance. You havent used it at all.
I think or guess that 98% of the liseners don't really care or know if it is a real singer or something else who sings. Is it possible to use all 80 singers for commercial need.
I know right? I wish there were AI products that would do the mundane household chores. Then again, there already is. We've got roombas, self-cleaning litter boxes, and dishwashers- but it still isn't enough!
I like your voice better than the AI, but I like the AI better than my voice! So this is a great tool for songwriters who have not (yet) developed their singing skills.
What an absolutely mind blowing presentation! However, you're wrong--you're still better than the AI. Your song at 7:00 is deep and expressive, while the AI rendition is excellent but not nearly as expressive. I would expect to hear the AI on the radio, while I would pay to see your live performance. Have you considered singing with an acoustic guitar instead of piano? I think it might be a phenomenal song.
One beautiful detail in your singing is that the F in the word "life" gets a little over-stressed (which is good) and sounds like V, which makes a great analogy: you never know where this life is gonna go and you never know where the words are gonna go either.
DEFINITELY, your voice better suited and expressed the song you wrote. The AI voice was very good, but just 'lacked' an authenticity factor (i.e. Emotion) AI was like listening to a recording. Your rendition was like hearing it live in person. More energy poured into the notation and words.
I appreciate the kind words and I can totally see where you're coming from. The emotion of a vocal performance is going to be extremely hard to replicate with AI. That being said, I'm sure it's only a matter of time before they figure it out. We will see! Either way it's mind blowing stuff.
The women voices were good, the male sounded to "modern", and i hate that sound. But with your singing ability you don't need AI. Me, on the othe side... I would like to hear the song example with you as lead and AI as harmony. I think that's how i would use it.
From what I've heard, the UI on Ace Studio far surpasses V-Synth. This is based purely off of the comments I've read- I have no personal experience with V-Synth.
I agree. I see this being mostly used as a creative tool to test out ideas and experiment. I could see using it to create vocal chops for EDM tracks, and using those in the final. But yeah, final vocals would almost always be the real deal!
Every ACE voice has a strong asian accent... And since Maxine, one of the founders, confirmed that they neither have any plan or idea on how to fix it, I cancelled the service. Yamaha Vocaloid, the product ACE copied, also has that problem in version 6. Perhaps they have fixed it for version 7. No one who isn't getting paid to do so, would consider ACE a useful product...
I do agree that most of the voices seem to have some degree of an asian accent. That being said, to say it is completely useless because of it is a bit of a stretch. I still think it's very useful!
I have already said this and I will say it again. Enjoy it while it last! The democratization of art/music by AI is cute but continuing unregulated, it will saturate the market to the point of making it uninteresting and the interest levels will drop to 5 seconds per song. In addition, musicians may well look for other occupations, perhaps carpenters or something like that since anyone can publish music even if they don't know anything about music!? Just wait and see. Stay safe.
Beautiful song you wrote for your child! The AI did NOT prevail, however. It's just not.... you. And for anyone invested in the song (your family and friends, for example)... they will pick your voice every time... millions of years of evolution have determined this. It's why AI may never capture our heart in all things art (or sports... or anything that collides with our animal instincts of procreation, competition, fear, shame, rage... etc.)
You're not wrong. That being said the AI voices are only going to get more and more convincing. At some point it'll be impossible to tell a difference. I'm super curious how long that will take? It'll be wild.
What will the money men do? I predict they'll stick with real singers for the lead vocal and as the genuine human face of a brand, but AI will be used to save money on backing singers.
It's going to be interesting to see! I think the big time producers will almost always use real singers because they have the budget and talent pool to utilize. That being said, I could be dead wrong. It's a wild time to be alive!
I agree, but I'm optimistic that these tools will used mostly by people who don't have the resources or ability to hire real singers. I hope it's used to empower creatives rather than replace talent. We'll see how it all turns out!
Congrats for your channel.
Useful, well produced, well edited, fun and informative.
I am an intermediate guitar player myself. Although I can play most of the parts of different instruments, I have a hard time putting them all together in sync.
Your videos already show a lot but could you please do a “live”, real time creation (no edits) of a song so we can watch your step-by-step full workflow?
@@brunorocha3718 it’s coming very soon! I’m going to start live streaming in 2025, and those will include sessions where I record a song in real time!
Welcome Teddy! I hope you and Mrs. Spencer are getting sleep. Oh yeah, the A.I, Love it. You changed my composing life, Spencer. Thank you!
Thanks for watching and I'm glad you're enjoying the content!
Teddy has been a fantastic sleeper. We got really lucky. He goes to bed every night around 9pm and sleeps all the way until 7am. I know that might change but I hope not!
you can also convert midi and lyrics directly form your recorded vocal in ACE Studio, just drag and drop the vocal audio on any singer track
This is by far the best video that I've seen that demonstrates what this AI tool can do. Like others, I think it's really good sounding as a backing vocal but not too sure about lead vocals. Also, it seems geared for pop. Do you know of any real implementations of rock vocals?
Audimee has some screamers. But it’s a converter so you’d need to create then convert probably.
Thank you so much!
I recently did another review of a service called "Audimee". It's a converter, so you'll need to record you own voice as a placeholder and then convert it to one of their rock voices. Audimee's rock vocals are super good, 100% recommend!
My opinion? I think it's a good way to demo songs, especially vocals, if the singer is having trouble, we can create midi vocals, then the singer has something to sing with while recording. Also if we create songs, we can show the ideas to our band that we have on mind. But something is missing, when the midi vocal sings the high parts, it doesn't sound right, sounds fake, that's why the real vocal will always show the emotions better.
They often call that a scratch track.
I agree, this is really great for that. An actual recording still feels and sounds better, but this is so great to convey an idea or example!
I agree that the emotion of a real performance is something that AI has yet to fully capture. It is a great tool to plot out ideas though. It's very empowering!
Hi Spencer, great video as usual ( congrats for your baby boy🥰) 👌 on the last song I preferred your voice ✌️ but interesting tool 👏
Always is not going to be the case for long. You can see it in another ai program like this called sythnv. Dont worry 3D artist know the feeling. :)
Try Synth V from Dreamtonics.
It sounds way better than this rubbish especially if you use the Solaria voice.
I liked your vocals brotha! Great video!
Thanks man, I appreciate it!
as for your question i think a blend between your voice and the AI's is the best way to go. I believe AI in the future will be utilized best that way. A hybrid basically, and if you can't sing. Well then perhaps work really hard on the AI to make it absolutely spot on! I love the vibrato tool it looks amazing, can you bend up the notes too? If not i'd suggest you tell the developers that since that is a common human way of singing.
Yes you can manipulate the pitch any way that you see fit. There's a ton of control inside the program- it can actually be a bit overwhelming at first. Once you get the hang of it it's pretty straightforward though.
This is both shocking and kind of fantastic. It's actually the UI/UX which has me most excited. Good voice cloning has been a thing for a while now, but it has never been so usable. What I would love to see is how this pairs up with one of those sing (or hum) to MIDI apps to generate notes, and then take that MIDI and send it into ACE. That would at least theoretically allow someone who can't sing to have something in their head and with some effort get to hear what it actually sounds like when it isn't being murdered.
The UI has come a long way and the level of customization is absolutely wild. I love that this tech is empowering people to make the music they hear in their heads- it's absolutely awesome!
💯💯💯
Spencer, I love that you constantly experiment with new technologies to stay ahead in our ever-changing world. Nothing matters more than relevancy. I appreciate your talent, passion, creativity, and refusal to relinquish your dream despite adversity. It is beautiful to watch and listen to!
Moreover, your vocals sound better than the AI vocals because I can instantly detect their robotic sounds. I have always enjoyed your singing voice and unique delivery. Although this software would be useful for a songwriter pitching a quality demo, especially in a female vocal dynamic, if the song calls for that and the user is a male vocalist or has no singing ability.
Combining your human vocals in Logic with the perfectly-pitched AI tracks panned opposite each other would make the finished song sound more realistic and better.
Thank you for introducing me to this new technology. I look forward to your thoughts on my insights!
This is my favorite RUclips channel, and I can see that your fans and followers are growing by the minute. Please don't ever stop doing what you do. You are helping so many people!
Your support means so much. Making these videos is a ton of work but I absolutely love making them, and comments like these are why I do it. I firmly believe that anybody can make music, and the fact that people are taking something away from my videos is incredibly rewarding. So thank you for sharing that, it made my day!
I do agree that the roboticness can be noticeable at times. I think if I spent a bit more time fiddling with it, it would be better. I agree that a hybrid approach is probably the way to go. I think recording a real vocal and using the AI for harmonies would give us the best result!
Wow! You have such sn Angel voice!❤
Your kind words mean a lot! Thank you
Hi Spencer, great video as usual and congrats for your baby boy 👌 For the last song I preferred your voice 👏 but it’s an interesting tool 😎
Thanks for watching and sharing your opinion! It's greatly appreciated.
Simply talented. Sana all😆
Wow. Very impressive! For your banger, I liked your performance better because it projected more energy and intensity at the high notes. The AI singer felt like holding back. This can probably be fixed with more editing in ACE studio. But both performances rocked!
The intensity is definitely something that needs work- it's hard to capture that with AI at this point in time.
So glad you showed the consonant sliders and other tweaking. Many videos make it look like you just type in the lyrics and you're good to go. You can spend a lot of time fiddling with the words. Deep voices are not as good and long sustained notes can be very iffy. That said, I'm having some good success making choir rehearsal tracks....would like some voices that are less pop or hip hop for choir.
Thanks! I agree, I'd like to see more versatile voices. Rock voices would be cool, as well as more classical style voices. I'm sure it's only a matter of time before they're available!
My answer to your question: of course your voice, Spencer!! Big soul, excellent talent, georgeous videos!! Thank you very much for all your hard work.
My best wishes for you and your beloved ones, from Spain ✨🤗✨
The continued support means the world to me. Thanks for watching!!
Thanks for new video - great as always!
Thanks!
That’s quite amazing
What a time to be alive!
Works absolutely great with your voice but then you're a very good singer 🙂. In this case I'd choose your lead combined with ACE harmonies, that opens up interesting perspectives regarding formants. In general I'd go a little easier on the vibrato in ACE, if you start doing it on every long note it might fast become unnatural. IMHO that is 🙂.
Can you sample your own voice for use in the app?
It is a powerful and very useful tool, but I do prefer your voice a million times over the AI examples. I would use it more as a demo creation tool for giving the melody guide to a singer for a project to let them know what I want for the song. The timbre, genuine feel, and overall performance could never be replaced by any AI no matter how good. Just like you said, this is just my humble opinion. By the way what a charming and lovely tune the one one you wrote for your son. Blessings for you and your family and thank you for such amazing content always!
Thanks for sharing your insights and the kind words! it means so much. I do agree at the very least, these AI tools are fantastic for developing ideas and creating demos. I love being able to whip up a woman's voice right from my computer, it's incredibly convenient and creatively empowering.
@@songsbyspencer yes Sir, I totally agree. Fantastic tools indeed!
Are you still subscribed to ACE Studio?
Try raising the Air parameter for the reduced airflow to make Naples super powerful, there's the rock/metal potential
I'll have to try this! Thank you so much for the tip.
your voice is much better👍👍👏👏
Awesome as always!
Thanks!
Is it possible to bye whole version so you can have it on HD?
What is the system requirements?
As far as I know it's only available as a separate application currently. I'm having trouble finding the system requirements unfortunately. If you have a decent computer (16gb of ram) and it isn't super old I would imagine it would run just fine.
@
I’ve got a answer
1. Compatibility with iMac M1:
Yes, an iMac with an M1 chip is more than sufficient to run ACE Studio smoothly. The M1 chip provides excellent performance and efficiency, so you should have a seamless experience while using the software.
2. Storage Requirements:
To ensure optimal performance, we recommend having at least 5 GB of free disk space for installation and smooth operation of ACE Studio.
3. System Requirements:
Operating System: macOS 11 or later
Processor: Apple M1 chip (or Intel-based processor with a minimum of 2 GHz)
RAM: At least 8 GB (16 GB or more is recommended for better performance)
Disk Space: 5 GB of available space for installation and operation
DAW Compatibility: Works with most major DAWs including Logic Pro, Ableton Live, and others
That great! Maybe for harmonies that are a bit out of my voice range it could be awesome! (Not sure for the lead vocals though)
I think a Hybrid would be a good approach. Use AI for harmonies while the lead vocals are real. It'd be pretty convincing no doubt
@ agree!
Beautiful and very authentic.
That ballad tho
INSANE!! It will help those how want to recorde a guide and don't sing that good, that is my situation 😂😂😂
Exactly. Vocals for everybody!
Amazing 🤩
Thank you so much for watching! The support means a ton.
I think I would love to use this if I'm strapped for time and want to add a lot of harmonies. This would prevent me from doing so many takes, BUT I would want it to be my voice (in most situations, although I am not opposed to other voices as well). Is there a voice upload/record option? For lead vocal, I would want it to be organically me. So I'll sing that one myself :)
The hybrid approach is the way to go! And yes, there's a way to upload voice data and create your own voice.
I prefer your voice because on the "Woah's" there was just a little bit of grit as you leaned into it. The Spencer voice had a touch of it, but not enough in my opinion.
Is there a way to dial that up with the AI voices?
BTW, loved the track itself!
They have settings you can adjust to raise the intensity. In my opinion, it isn't the best at doing gritty vocals. I hope they update it!
Hello! Can it sing in other languages, like Portuguese?
As of now it only supports English, Spanish, Chinese, and Japanese. I'm sure they'll add more languages eventually!
This is just Synthesizer V by Dreamtonics. The solaria voice is nice and there are more 'tuning' options like breathiness, or belt sounds
It's cheaper. Synth v is greedy and cost 79 a voice. I uninstalled synth v and chose this one.
@ChongFreak as VSTs go, 79 pounds is cheap in comparison and judging solely on this video much better results and control
Spencer, your voice is clearly better!!
I don't know the technical terms to use, but it sounds like your high notes are more "shiny". Perhaps it's the human placing the voice slightly differently compared to the AI voice.
@@johnnyfilippone Spencer didn't really properly used the energy, tension and note bends on the ai to simulate that high end so its not a fair contest. I been using it and you can't just spam notes. You gotta start from a down note and take advantage of its capabilities to realistically pitch bends and energetically raise at certain parts of the singing.
Naples doesn't really have the "shiny" bright tone to start with, although that makes him in a weird position for where he's still best suited for the 4th octave
source: am an ace studio user
This sounds great. I have tried to find some Ace Studio examples of rock vocals to use in my productions. Does Ace Studios have rock vocal examples, and if so, can you point me in the right direction?
I haven't had the best luck creating Rock Vocals with Ace Studio. Most of their voices are best suited for Pop. I'd like to see some Rock Voices added. Hopefully they do so in the future!
Thanks for responding.
i love it
Thanks!
I am an amateur. I think Ace studio is great, but I need to create songs in Greek. Waiting for this expansion..
I hope they add more languages soon!
Hi, I’ve seen many demos of this ai on YT and yours is clearly above all the others ! Congrats !
Thank you so much! I put a lot of time into this and I wanted to make sure it was good- comments like this make all that effort worth it.
Yamaha's VOCALOID has been a legend in this field since 2003. The difference is that back then, the AI tag was not used, whereas nowadays, the AI tag is prominently associated with it. VOCALOID 6 can also be used as a VST plugin in your DAW.
I'll have to check it out!
I used Vocaloid a few times as examples for real singers to have an idea of what I was aiming for and for that it worked wonders but Vocaloid is "miles" away of this new AI stuff.
Bs. I used it. No good. This one's really good. The conversion is absolutely gorgeous.
Vocaloid doesn't sound anything like a real vocal. Someone who is very creative can make it sound decent, but it cannot sound realistic.
I still hear some areas that aren't quite convincing. But this software seems to come with enough tools to tweak it until it does sound more realistic. It's certainly already leaps and bounds above Emvoice, which I've been using a bit for backing vocals. That being said, I suspect we're really not that far away from truly convincing AI vocal tools. Not sure if that's a good thing or not.
I agree that there are some parts that could be more convincing, but like you said- the tools are mostly there. It's going to be absolutely wild to see where this technology is 5-10 years from now. It's super scary yet exciting
@@songsbyspencer Yeah. It kinda reminds me of how, 25 years ago, I was convinced that there would never be a guitar emulation that would sound authentic. Which sucked because I can only play keys. About 5 years ago, I decided to Google it and see if there was anything half decent. I was blown away! Now I've got several guitar/bass/drum VSTs, and I started working on a bunch of ideas that sat on the back burner for years.
Nice!
Thanks, dude!
Does it handle Brazilian Portuguese? And can we make a custom singer out of our own voice?
You can create your own voices by uploading data and modeling it- so that's pretty sweet! As of now they support English, Spanish, Chinese, and Japanese- so no Portuguese just yet. I'm sure it's only a matter of time.
@@songsbyspencerMuito bacana da sua parte separar um tempo para responder a todas as perguntas, ainda mais levando em conta que você tem uma criança pequena para cuidar.
@@songsbyspencer Thanks.
Nicely done. Consider blending your singing with that of AI.
This needs to be available as a plugin for DAWs, rather than a stand alone app that requires all that importing/exporting. That's not the optimal workflow if you're creating and experimenting
I agree, it would make workflow so much easier. I'm sure we'll see this tech in plugin form soon!
It does work inside a DAW via a VST plugin. It's called ace Bridge I believe and it is located in the Ace folder. Just copy the dll and paste it in the vst folder that your DAW is pointed to for vsts. I use Reaper and this worked for me.
AI can't even come close to having a nice "growl" like yours, Spenz 12:01 😁... but this ACE Studio would sure give Dreamtonics Synth V a run for their money 😆 great demo from you btw
Thanks for watching and the kind words!
@@songsbyspencer thanks for replying... I always hit the sub button when an uploader replies on my comment... new subscriber here 😁
Amazing program bro! Can I turn this program an 80's metal singer? I already use A.I. to make music for the Bible. Later (I hope) when I will have more free time I would make tons of songs in 80's metal styles.
I didn't see many rock-centric voice models on the platform. That being said, I'm sure it's only a matter of time before they're on there.
Are you able to edit your midi notes while listening to the vocal in the context of the rest of your track?
If you export the audio of your track, and pull that into Ace Studio, you could listen to the music behind the vocal being generated, and edit as needed.
Anyone know how well this works with Cubase? I heard of some "bridge" thingy but not many vidoes on it? Anyone?
Can you upload your own rvc models?
As far as I know, yes. You can import your own voice data to create your own voices!
It’s a step in the right (or wrong) direction. But I think the voices available are very mediocre at best. But if they got decent vocalists, then it could be a game changer if it works as good as you’re portraying in this video.
Your voice still is better, but not for long I guess. Best regards from Portugal.
Thanks for watching! It'll be interesting to see where this technology is in 5 years.
Great A-B comparison with your human voice vs ai “Naples”
imo… human voice superior.
For how long? who knows? 🎶🎤
Give it 5 years and the AI will be completely undetectable!
AI voice changed based on your recording would work so much better as you have some really nice inflections and tonal slides that is not captured in the MIDI, making it sound less natural and realistic. Great demo though!
In the video I plug at the end of this one, I review Audimee- which does exactly that! It transforms your vocal takes into different voices, but maintains the same characteristics. It's awesome!
The female voices on this platform seem stronger than the male. The other AI site you reviewed had great male vocals. For the “banger” I liked your vocals better, but want this site does really well is the vocal harmonies….when the voices are layered they sound even better…
I agree! As with all AI tools I think they're best used when they complement the real thing. This has been super useful when layered in the background!
for slow songs nice but whats with 200-300bpm songs?
I would imagine it would work just fine. 200-300bpm is hella fast!
@@songsbyspencer oh really this works? good to know. yes 300 is really fast :D
Can ACE Studio import musicxml files (exported by MuseScore), which already contain the lyrics?
I've got loads of choire sheets I'd like to vocalize.
From the quick search I just did, it says you can!
@@songsbyspencer Yes, I tried it out. It works. Not with musicxml, but with MIDI. Lyrics get exported in MIDI. I didnt know that up to now. Thanks for the reply!
I would totally use this for demos. I would not production release this with AI singing lead or even backing vocals because the service that does this can potentially file copyright claims for the use of their product. Don’t like the legal implications for this for anything other than personal or demo use. Also, who’s going to buy an album knowing that it was AI-generated? That cheapens the value of YOUR music. Demo? Yes. Production release? Hard no.
Very fair point. I really hope AI is used as a creative tool as opposed to making artists obsolete. I could see it going either way. You raise a great point though, I don't think people will be interested in listening to AI music. As a novelty, sure. But listening to an hour long album that's entirely AI would put a lot of people off.
QUESTION: at 8:43 can't you just DUPLICATE the layer and not have to type all the lyrics all over again?
That seems like the easiest feature for the computer to do.
I wonder how it sounds with some good music.
Hater's gone' hate!
@songsbyspencer I really do not hate this kind of music. I just do not find it good.
Your voice is waaaay better! But....in about 12 months I'm not sure.😊
Yeah no kidding! The pace the AI is evolving is absolutely insane. I can't even imagine where we're gonna be in 10 years.
i prefer your voice of course, but i use Ace studio too (im not a good singer)
I appreciate that! I"m glad you're using it and I hope it's been useful for you. I really like being able to generate female vocals, that's my favorite part. Emma is my favorite voice!
Respect to the A.I., but I think your version would be my pick!
Thanks!
V great interface. Voices in synthesizer v are way better.
I love your videos and I dig the channel, but man... to me this AI stuff is a bummer. I get it; the tools are out there and I guess someone has to showcase them, but it doesn't do anything for me, at all.
Also, you just wrote lyrics and melody but uploaded it to the machine for it to sing: Who owns the idea now? Genuine question.
Anyway, happy holidays and all the best.
I can totally see where you're coming from. This technology is absolutely incredible, but it's also scary. I see the angle of how AI might devalue a real producer's work, but I also see the angle of how it lowers the barrier of entry and gets more people to start making music. I'm optimistic and I'm not overly concerned with AI taking over music production. Creating great music requires a lot of creativity, and it's an extremely nuanced process. If someone were to only use AI tools and premade loops to create their song, it isn't going to stand up against a producer who creates each element of their song by hand. A real producer feels an instinctual need to make music, and they will do so with whatever tools they have available. That passion always shines through, and it's something I have yet to see AI come close to replicating on a large scale. It can make pretty convincing vocals if you tell it what to say, what melody to follow, and you work with it. But for it to come up with all of that on it's own? I have yet to see it. But who knows, 5 years from now I might eat those words!
Also, the creator of the lyrics and melody is the writer of the song by definition, so I don't think there are any legal issues there. I know most of the voices in Ace Studio don't require a license and are royalty-free. There are a couple voices that require additional licenses to use, but they're clearly marked to avoid confusion.
can´t you make AI use your voice? And play around with it?
You can upload your own voice data and make your own voices. So in theory, yes!
"I get that feelin' Woah!
Makes me wanna say Oh Yeah."
Did AI generate those lyrics? LOL.
No- I was trying to write a modern pop chorus, so I really dumbed it down! LOL
Atm monthly subscription is U.S.$22/month, and the annual one is U.S.$149/year. That will stop the hobbyist, but so it was with Ai-image creators, so .. A 'week' from now, and it free..
The 'battle' Imo a draw, but exact at the beginning Spensers voice had more ..i dont know *appeal* i guess it could be called
Synthesizer V Pro needs to step up their game. This UI is so much better.
Competition breeds innovation!
Can the vocals be edited further to avoid some of the funny flutter?
Of course! You can tweak it until it sounds right to you. There's a ton of tools inside and they give you a tremendous amount of control!
@ thanks for the reply! Your videos are great!!!!
This is how AI could be used for composers that can't sing at all. I wish i had this as a kid, imagine the kids growing up with this tech. Please kids, make amazing music again. Now the dreams the possibilities are endless, the only thing threatening your amazing songs is the record labels, but hopefully they'll lose against the AI companies. AI is doing imitations just as we humans do. When i play guitar i try to imitate Gary Moore all the time should i be brought to court now? Interesting times to be alive!
This is a very refreshing, positive, and healthy attitude- and I 100% agree. These tools are absolutely awesome and empowering!
Naples for sure
He's got those sexy Weeknd vibes. I can't compete with that!
Can it do different lenguange?
Yes, It can do English, Spanish, Japanese, and Chinese.
@songsbyspencer awesome!
Spencer you didnt really use the voice editor to the max. It can pitch bend and slide up notes but you did it word for word and straight note key for key which made it sound ai. Your vocals you were using pitch bends and all, you were pitch shifting and doing all the emotions which you didnt edit into the ai. It shouldnt be completely 100% note for note is what i am saying on the ai edit. Theres a lot of different notes in 1 note. Ill send you an example when i have a chance. You havent used it at all.
I think or guess that 98% of the liseners don't really care or know if it is a real singer or something else who sings. Is it possible to use all 80 singers for commercial need.
Great! All creativity is now AI-powered! So we can go do laundry and wash dishes
I know right? I wish there were AI products that would do the mundane household chores. Then again, there already is. We've got roombas, self-cleaning litter boxes, and dishwashers- but it still isn't enough!
Emma sounded real. The rest sounded autotuned above their natural range.
Emma is by far my favorite voice on the platform!
@songsbyspencer Glad it's not just me.
OK, your voice sounded better to me on the last song.
I appreciate that!
I like your voice better than the AI, but I like the AI better than my voice! So this is a great tool for songwriters who have not (yet) developed their singing skills.
I appreciate that! And yes, this is a great tool to work out ideas and experiment without having to rely on your own singing skills.
I'd buy it for home song writing if it had rock / metal voices and it wasn't a rip off subscription model.
I'd like too see some rock voices as well. All the voices I saw were geared towards pop. I'm sure they'll have rock and metal voices soon enough!
What an absolutely mind blowing presentation! However, you're wrong--you're still better than the AI. Your song at 7:00 is deep and expressive, while the AI rendition is excellent but not nearly as expressive. I would expect to hear the AI on the radio, while I would pay to see your live performance. Have you considered singing with an acoustic guitar instead of piano? I think it might be a phenomenal song.
One beautiful detail in your singing is that the F in the word "life" gets a little over-stressed (which is good) and sounds like V, which makes a great analogy: you never know where this life is gonna go and you never know where the words are gonna go either.
The song was actually originally on guitar. I just switched it to piano since the other samples were on piano!
That's because he kept it simple. You can make it expressive using tension, and many more but it takes time. This ai is very capable.
You are better than Naples
Ya hear that, Naples? I'm comin' for you!
Thanks for watching- it means a lot!
DEFINITELY, your voice better suited and expressed the song you wrote. The AI voice was very good, but just 'lacked' an authenticity factor (i.e. Emotion)
AI was like listening to a recording. Your rendition was like hearing it live in person. More energy poured into the notation and words.
I appreciate the kind words and I can totally see where you're coming from. The emotion of a vocal performance is going to be extremely hard to replicate with AI. That being said, I'm sure it's only a matter of time before they figure it out. We will see! Either way it's mind blowing stuff.
It seems a very good plugin but so expensive. I'm a hobbyist not a star.
thats insane for peapole records songs but dont sings like me
Exactly! it's a complete game changer
is there a free alternative to ACE studio?
The women voices were good, the male sounded to "modern", and i hate that sound. But with your singing ability you don't need AI. Me, on the othe side... I would like to hear the song example with you as lead and AI as harmony. I think that's how i would use it.
The consensus seems to be that a hybrid approach would work best. I'll have to give that a try!
Twice the cost to own as the pro version of V-Synth with Solaria.
From what I've heard, the UI on Ace Studio far surpasses V-Synth. This is based purely off of the comments I've read- I have no personal experience with V-Synth.
Only twice? How about in 6 times? Ace studio pro is in around 6 times higher price if you want own.
A good tool for writing, but prefer a real recorded vocal, even if it's technically imperfect.
I agree. I see this being mostly used as a creative tool to test out ideas and experiment. I could see using it to create vocal chops for EDM tracks, and using those in the final. But yeah, final vocals would almost always be the real deal!
Every ACE voice has a strong asian accent... And since Maxine, one of the founders, confirmed that they neither have any plan or idea on how to fix it, I cancelled the service. Yamaha Vocaloid, the product ACE copied, also has that problem in version 6. Perhaps they have fixed it for version 7. No one who isn't getting paid to do so, would consider ACE a useful product...
I do agree that most of the voices seem to have some degree of an asian accent. That being said, to say it is completely useless because of it is a bit of a stretch. I still think it's very useful!
You can upload your own voice as a model.
Your Voice is better than Naples.
Much appreciated!
I have already said this and I will say it again. Enjoy it while it last! The democratization of art/music by AI is cute but continuing unregulated, it will saturate the market to the point of making it uninteresting and the interest levels will drop to 5 seconds per song.
In addition, musicians may well look for other occupations, perhaps carpenters or something like that since anyone can publish music even if they don't know anything about music!? Just wait and see. Stay safe.
I really hope you're wrong- but you're probably right.
@@songsbyspencer Believe me! This is one of those few times that I do really hope and wish to be wrong.
There will always be The Beatles.
Beautiful song you wrote for your child! The AI did NOT prevail, however. It's just not.... you. And for anyone invested in the song (your family and friends, for example)... they will pick your voice every time... millions of years of evolution have determined this. It's why AI may never capture our heart in all things art (or sports... or anything that collides with our animal instincts of procreation, competition, fear, shame, rage... etc.)
I prefer your singing to the AI voice. The AI voice just didn't seem real or emotional enough for me
I could see it
It´s not free..........
😮
$432 over two years to own: pretty steep pricing for essentially a plug in.
How much does Microsoft charge for their junk?
Sounds fake but good tool to overlay over real vocals to get a fuller sound.
AI digital "clone" voices sound often boring and lame, no emotions, no real feelings.
You're not wrong. That being said the AI voices are only going to get more and more convincing. At some point it'll be impossible to tell a difference. I'm super curious how long that will take? It'll be wild.
What will the money men do? I predict they'll stick with real singers for the lead vocal and as the genuine human face of a brand, but AI will be used to save money on backing singers.
It's going to be interesting to see! I think the big time producers will almost always use real singers because they have the budget and talent pool to utilize. That being said, I could be dead wrong. It's a wild time to be alive!
Not a big fan of A.I. potentially replacing singers.
I agree, but I'm optimistic that these tools will used mostly by people who don't have the resources or ability to hire real singers. I hope it's used to empower creatives rather than replace talent. We'll see how it all turns out!
first here to comment ❤❤❤😊
Without actually commenting. Doofus.
That was a great video, very interesting tech on display. The world is changing. Thanks for posting.
The support means so much
Thanks for watching!
The vibrato sounds intensely fake, similar to excessive autotune.