Bro, this is a great video. You really perfectly blended what is essentially a gear demo/product placement with a practical application and a lesson in how to present the best possible version of an idea regardless of technical limitations. I’m a crappy singer who communicates toplines for my “backing tracks for the song I forgot to write” with VSTs and such, and this totally nails the target audience (with my sample size of one 😂)
i see it as he was saying, a tool for demos and what not. at the end of the day he'll want an actual singer to sing it. i always felt the competitions were more about final products.
@@CreativeMindsAudio I love how literally no one reads the info about the contests before making comments on it. You have half a point, but no, it's not about a finished product. It's about a finished SONG. AI is obviously strongly discouraged because a song is about the performance, and AI can't "perform" per se. It requires absolutely no skill or artistic effort. But the production can be FAR from finished and still land as a valid submission. You can have a finished song and it be not even close to what you'd hear on a release, and AI definitely isn't something a person in their right mind would put into a finished song, except... MAYBE... Eleven Labs voice modding, but that's still a real singer, just with a drastic timbre and tonality shift.
Ugh, it's actually affordable. I wonder how well you could use it to fix some of Suno/Udio's mistakes (if inpainting doesn't fix them), like extract the vocals replace them with this?
Really great demo video of this. Well done. Just to clarify a little off topic for your sake as well as other viewers...as Songwriter in Nashville of 16 years interacting with Publishers, writing and pitching songs and studying with numerous Hit writers...it's the general consensus that the vast majority of Female Artists actually PREFER Male vocals on songs getting pitched to them. Reba especially but most others as well. They often don't want a vocal that leads them too closely in artistic expression as they want to add that of their own. Just an FYI :)
That's pretty interesting to know. Even if it's way outside their range? Like they'd rather hear me falsetto it? Or I should sing it in a good key for my voice and they'll put it in whatever key?
If lazyness has always been a very serious problem in your life, to the point of turning your life into a mess, i strongly suggest that you take some time to study about ADHD (Attention Déficit Hyperactive Disorder). Although it has the "Hyperactive" word in there, it's important to know that this disorder CAN destroy the person's ability to feel motivated or to sustain motivation, for this reason, a LOT of people with ADHD are HYPOactive instead of being hyperactive. People with severe ADHD that suffers from severe lack of motivation are ALWAYS labeled as LAZY. They themselves grow up believing they are lazy and feel a lot of guilt and shame.
So, they are filling the gaps one by one If you can't play instrument but can sing use this tool If you can play guutar but not piano or drums use this If you can't sing but play an instrument use that If you can't compose or write a lyric or mix or master use this or that Feeling me now It's only a matter of time before it throws everything away including us as well 😮
There have been pedals on the market that have done that for a while now, like TC Helicon's Play Acoustic. Voice tune, pitch tune, octave change, gender change, accompany, harmonics, etc.
I have nothing against sponsored videos, when it is disclosed as you did. The problem in this case, I think, is that it prevents you from informing your viewers about the many options available to do this. I prefer Synth V, which is remarkably similar, but is not subscription model. Others who can sing reasonably well might prefer “voice changers” such as Audimee (there are others). Thanks for your video, anyway.
I used to try to use similar tools like SynthV and Vocaloid before I got better at singing. My biggest complaint with tools like these is that they almost always sing with so little power and attitude/personality, which works for a lot of genres but not quite so well with rock or metal oriented music. I'll still occasionally use it as a writing tool, but it can be hard to really conceptualize what the vocals would sound like with a proper belting singer sometimes
VocalSynths can sing with a lot of emotion, you just have to know how to do it and have the right voicebank. I feel UTAUloids are better about having emotion, but again, depends on your ability and the voicebank
So you draw a firm line between this and the generative AI being “cheating”… So what if it’s being prompted using my original music and lyrics? I realize this video is a couple of months old already and that’s ages in the AI world. But I’ve been using Suno exclusively for vocal tracks to use this way. No, I can’t input the exact midi notes or the timing but that’s what makes it so impressive. It understands what to do with it even better than a real vocalist would. A certain number of prompts will give me ideas that don’t sound like what I want for my song and even if they sound pretty good, I delete them and move on until I get something that is true to my creative vision. Just like yourself, I would love to use these as demos to record with a real vocalist. But my aspirations as a musician aren’t focused around recording professionally or putting a band together. I just want to finish songs I started 20 years ago. And anyone who calls it “cheating” can just suck my balls 😊
i really like the concept so I took a look at their prices. If someone is a full-time songwriter who can't sing, this would be an excellent tool at a great price. I'm a little concerned that the price kinda puts it out of reach for the average "in my spare time" musician, though. That's about as much as some plugin suites for something you might only use a few times a year.
Hi.. Does the AI signer work as a creator of a vocal lines and melodies? Like if I have song without vocal parts, I'd would upload my song into the AI and it will create a vocal lines?
It's still got some odd sounding artefacts and unnatural decay on the ends of notes, making them sound clipped, but in 10 years this technology will probably be amazing. To do extreme metal vocals, one almost has to be a cartoon voice actor, with various ranges and timbres. If they can do Cradle of Filth style vocals with voices/clean singing/fry screams etc., it will be something. It isn't just about the notes. It's about tweaking vocal weight and timbre. Right now this seems really tedious to get right. Like tweaking velocity/timing settings on hi hats for a more human sound.
Two things: 1) In my experience, there is NO better way to give yourself a solid, intuitive sense of how to write vocal melodies than taking voice lessons. As in, within my first month and a half, my vocal melody writing became ten times better than it was before lessons. 2) BOY OH BOY, would this tool have been life-alteringly helpful in the days before I took voice lessons. Holy shit.
Yeah, I've had people send in songs that were written with AI voices but it sounds like they've never heard a person sing before, like they wrote th melody on guitar and it's something no person would ever sing
Since it converts to midi, is the quality of the microphone / recording out the window too? I assume that is the case similar to drum performances converted to midi are only looking for the hits and possibly velocity
It means that it doesn’t matter what microphone YOU’RE recording on, it’s always going to sound like the one THEY used to record the voice they trained the AI model on.
Yo man! Did you try SynthV before? I think SynthV can create a more realistic voice and it is a life time license. ACEstudio subscription is pretty expensive man😅
im against most ai use but this seems fine to use as a last resort. im just some guy in my basement and if i wanted a female singer, i dont have the resources to get that in the first place. i think it depends on your situation. if you can do it for real, do it 100%, but if you cant, i dont think its that big a deal.
Not gonna lie, this software reminds me a lot about Vocaloid (using a piano roll and phonetics, amongst changing vibrato and other properties), but on strong steroids.
I have a good singing voice but have issues singing without hearing backing vocals. Does anyone have any suggestions for getting better without them? I know this is more a guitar and songwriting channel. I think honestly its just practicing singing to instrumentals
Backing vocals ad lib over the main vocals. A rapper with a hype person could have a ding in the head phones to note when to start and start again. Adding and then taking out the sound markers shouldn't be a problem for a sound engineer using multi-track.
:40 - IMO this is a false dilemma. Even if the primary tool is generative AI (such as Suno, Udio), there is still tremendous power within the interface (intricate prompting, lyric input, stem exports, etc.), and this will only grow more in the future. There's also an aesthetic evaluation by the user that comes into play as outputs are culled. Now, if ALL one is doing is typing in something like "catchy pop song" into the AI and accepting the output and sharing it, we're in total agreement there. Maybe I'm optimistic, but I think listeners will soon be actively listening for such work, and weeding out a lot of it. Good channel, I should say.
Seems like a great tool but not sure if I'd get much done as I'd waste hours punching in the darkest lyrics possible, including all the "RAWR"s, "BLEGH"s and "ARF ARF"s, but use the sweetest angel voice available, just for giggles.
Very Cool! Now if We could get something a little More Metal ala Rob Halford ,Bruce Dickinson for some good Power Metal, maybe King Diamond. or just go for Guttural death Metal stuff. still though. like I said. Very Cool!
One problem with this is that some “creators” will undoubtedly release songs with these AI vocals on the finished product, possibly as the only top line. Your expressed use for this tech makes sense, but people will not adhere to it. Social disrespect for this choice will be unsufficient to stop AI voices / top lines from becoming normal. What happens in 3-5 years when this technology’s update or successor provides an option to write melodies as well? AI will be used to “compose” every aspect of music that we consider intel property. I love what you do Trey, but it’s irresponsible to promote such tech developments.
There's already loads of AI that does that kind of writing, and it's terrible, but this isn't that. Just because lazy people will abuse this and put out music with this as the topline doesn't mean it shouldn't be promoted as an incredibly useful tool. Lazy and uncreative idiots will always abuse any new tech.
@@treyxaviermusic I hope you’re right. Just to balance out my negative comment, I will admit that you are, IMO, the most inspiring musician and teacher around. I look up to your musical knowledge and opinions more than those of anyone else.
Hopefully in that case, that lame arse music gets the attention it deserves, i/e ZERO. GENERATIVE A.I means you get fuckin zero respect and thats the way it should be, forever.
@@treyxaviermusic Actually I've just recently heard a few AI metal songs and they were VERY good! 😮😭 I mean, there was some hiccups vocally, but the riffs and melodies were there! And the reality is there's just SO much (real) music out there these days that it's hard to stand out from the sea of genericness. So, when u DO have a good riff AND good melodies u DO stand out! Thus these songs were some of the best I've heard in awhile and I hate them for it! 😫
It's always a complex subject. A tool Vs an ai that makes things to you. Respect man, but, if u can't sing and the aí sing to u, is the second type of ai. U know how a good singer sings, but u can't. A person can know how a good drummer plays, but can't do it. I think in the future we will have more "producers" than musicians. 😁. Complex subject.
Good point. Ai tools are exponentially getting better by day. Very exciting. Can’t wait for a AI co-producer / mixing engineer to make productions with. “More bongo’s jives and a new vocalist… yes please”
Hear and look the Video from the Recordingsession from "We Are the World, we are the Children" produced from Quincy Johnes, performed from •Lionel Richie • Stevie Wonder • Paul Simon • Kenny Rogers • James Ingram • Tina Turner • Billy Joel • Michael Jackson • Diana Ross • Dionne Warwick • Willie Nelson • Al Jarreau • Bruce Springsteen • Kenny Loggins • Steve Perry • Daryl Hall • Huey Lewis • Cyndi Lauper • Kim Carnes • Bob Dylan • Ray Charles and many others in the Choir. After you have hear this, you will know what it mean to sing Musik with all your Soul, all your Emotions, all your Energy and how it sounds, if realy skilled and trained Singers, real Professionals who are real Artists, performing real Music !!!!!!!! All this AI faked "singing" is crap for the trash !!!!!!!! AND THE LAST WHAT WE NEED IS A OTHER SUBSCRIPTION !!!!!!!!! SUBSCRIPTIONS ARE FOR PEOPLES WHO HAVE A MONEY SHITING DONKEY AT HOME... ...BUT MOST PEOPLE HAVE NOT SUCH A DONKEY AT HOME...
There are plenty of these tools. I just wonder why anyone would need an AI singer. AI doesn’t have an artist persona. You can't perform live with AI singing, and you can’t create any content with AI singing other than music without real people behind it. Who would be interested in that?
He's obviously being paid to say how good this is. The tech does show great promise, but the free version isn't very good, and doesn't do all he says it does.
Irritated by the start, I later on felt like saying thank you for making me - at least feel like about to - understand what's so relevant about AI singing: if you find a melody you can have the best or the worst to sing it. I would like to add - having seen all - that the presentation, the performance of any song is decisive. Like, quote, Tina Turner, when asked to sing a song: "Yes, I can - do it", i.e. make a melody a song. In fact, same thing, said from the other side. What you emphasize right from the start, to me, seems that from now on any tune or combination of tones can be "turned out" a song, if it fits the right voice. Also, singing - and only singers would disagree, at all - is clearly defined as a skill just like any other. Question: don't you need great voices for the input, from the start? The example you give at the beginning of your video looks like an argument.
I so totally agree to this! How in the name of our Lord and saviour were people in the past able to convey other people how to sing without AI! Man, can you imagine the songs we could have had! Forget all the hits of the past, we need to start over! With AI! I'm not taking one subscription, I'm taking three! My music will be so good and singers will love and adore me now!
DON'T KID YOURSELF PAL, tell that story to a professional singer,instead of justifying what you are doing. AI is what it is. It's acceptable for you to replace a singer because You are not a singer. If you see you wouldn't have that flippant attitude. When music production is replaced, come back and let's see what your attitude is about “musical integrity". AI is great until You loose your clients.
I mean obviously this is a great tool, but you just know some people are going to use it to cheat and it destroys my faith in music not to know whether I'm listening to AI or not. That's why these things suck.
That's the large-scale problem with AI though, and I very much agree with you on that. People always abuse technology, it's why we can't have nice things. But we can't let those assholes ruin everything
i can sing, but i don't like my voice. so i sing and record it and then swap the voices, i don't think this is cheating. ok, the people who can't actually sing would cheat then, but it would be a nice tool to compose, so its still human working on tracks.
It won't be a problem. The greatest artists of the future will use these tools too, and we'll think: "Daaamn, i have access to the same tools this artist has but i can't create amazing songs like him/her." In reality it will be good for all of us... because this technology won't make great artists worse, in a matter of fact it has the potential to help them make more and better songs. At the same time it will benefit a LOT of people that struggles too much to finish the songs they started creating. At the end of the day we'll discover a lot more songs that we like (and a lot more that we hate, but these we can ignore after we tried it)
@@brunomartinello1114 I disagree. I'm a songwriter and musician, and I can't think of anything I want less than to have something write part of my music for me. If any great-sounding music arises from AI, it will be soulless by it's very definition.
Yeah, OK so if I have a beautiful song sung by a male singer, you telling me a female vocalist will have no idea how to sing it unless she has a female on the demo come on, dude that’s ridiculous
No that's not what I'm saying - it's about presenting it in the best possible way to sell them on it. Obviously she can hear my terrible falsetto version and sing the same thing but better - but will she WANT to?
@@treyxaviermusic I did the thing man, I’m sorry. I reacted before watching the whole video. Anyway, I do agree that AI can be a great tool for figuring out concepts, as long as it isn’t used AS your final product. Unfortunately, too many people will use it as such. Keep rockin’ bud! 🤘
I would caution all on use of AI tools for vocals AND instrumentals, like this and Vocoflex, to use them only for prototyping if you care about reputation as a legit human artist. Vocoflex imprints their output stream with an indelible watermark; others, if not all will follow. If anyone declares that your music is AI and then checks for the watermark, they could then claim "you're nothing more than an AI artist! 😡", even if you used the AI tool as... only a tool... as you would any other production tool. However, in the case of AI tools, people will be quick to slander you as an "AI artist" and get others to dismiss you on that account. Don't get me wrong... I LOVE AI. In fact, I work in the field and don't see any way forward but to embrace AI. But that's because I am pro AI. Most people won't be, especially as job loss due to AI increases to a fever pitch. Be careful of how and when you use these tools, especially when publishing. Just an early warning.
first off i have issues with the crap you and other "SOCIAL MEDIA INFLUENCER MUSICIANS" spew the whole B.S. of constantly put new stuff out don't worry that it's not perfect. Me i am one of those true musicians who believe you do not release stuff till it's absolute perfect and meets your vision 100 percent i started playing guitar and writing songs in 1990 and i have over 3000 songs, or should i say song ideas since they are partial lyrics or just titles, since my playing skills are not up to meet how i want the songs to be. I will not take the lazy way out of using plug ins, studio musicians, drum loops, click, kick tracks, backing music tracks, or cut/copy paste instead of the old school way of double tracking, layering and recording parts buy actually playing the parts over and over again. ( wont go into why that is better and you get more out of that than using the lazy cut/copy and paste. also Modern day musicians/artists/bands have become too lazy and dependent on technology. MY motto is never put trust or faith in technology cause it will fail and or betray you when you need it the most. As for AI for singing. I can not sing my way out of a wet paper bag. My voice when i sing sounds like music played on a cassette that the cassette player is slowly dying and the music is slowing down with the vocals. That is how my voice sounds.I have a vision of what my music is supposed to sound like and I will not compromise it even if it takes me till i am 99 years old to release my first song/album. Those who use these lazy ways of doing Music whether it is click, kick tracks, drum loops, backing tracks for the music, Singing, Musical instrument parts (I have no problem with backing tracks for sound effects like crowd noise, gunfire, lightning, thunder, or narrative speaking like the prayer in Metallica's enter sandman or at the beginning of Megadeth's Captive honour track from countdown to the extinction) Then there is the lazy vocalist vocals, which is known as cookie monster vocals, or i call it either angry dog barking vocals or more accurate the lazy vocalist vocals, you know them as growl vocals. then there is the lazy using plug ins, computers, drum loops, instead of real instruments, real amps, real speakers, real speaker cabinets. I am tired of the whiny safe space/cry closet woke , entitlement snowflakes whining about it costs too much to travel with real amps, cabinets, rack effects, blah-blah -blah. Guess what that is a part of being a musician DEAL WITH IT. I will never to the falling in reverse Ronnie Radke thing. Me and my future bands will have real amps, real cabinets and real speakers, racks and pedalboards, and since my future band will have 2 drummers and 3 guitarists the drumkits will not be these what i call kiddie drum kits, They will be full oll out kits, like Neil peart and Terry Bozzio, etc. Also will not allow producers no matter how famous they are, do what i call the Dann Huff Move (you know without band, artists knowledge or consent, either they or a studio musicians they hired changes and or adds parts and or remove parts behind the artists/bands backs, as Dann Huff did with Megadeth) I will not tolerate that from any producer no matter how famous, if they did that i and i found out they would be fired on livestream so the world could see it and they would not get paid for their services cause they would have broken the agreement between producer and client. Right now todays musicians, artists have become too lazy and dependent on technology. Dave mustaine was 1000 percent correct when he said something to the effect that due to technology there is a lot talentless people recording and releasing crap and passing it off as music today. Guess what when you flood the market with crap, guess what happens, consumers start to buy less or stop buying all together. That is why the album sells, cd sells are way down. also bands, artists bought into the lie from record labels, social media clowns or whoever and that if you don't release new content, material every 3 to 9 months, then people will forget about you and move on to the next flavor of the moment. I can easily prove that to be b.S. Guns n roses Chinese Democracy album took 8 years and multiple lineup changes and when it was releases there still was tons of fans buying in and going to see them in concert. Putting crap out every 3 to 9 months hurts the quality of that material. You need to spend more time on the music. Otherwise you get filler songs. when people did buy albums, and artists, bands spent like a couple years to release music. You make get 1-3 and very rarely 4 filler songs out of the standard 10/11 track album. (yes i consider 10 or 11 tracks is standard for albums) today you rarely get 4 good songs and even rarer is getting hits. most of the songs on albums today are filler songs. they are as crappy as these left wing reboots of classic movies and tv shows. Me my sole passion is music and i will never compromise my vison for my music even if it means i will be in my 90s before releasing my first song or albums. other musicians, bands, artists, and "Social Media Influcencer Musicians" who are a lower class of musicians. May have no problem with releasing crap, constantly till they get to the jems. I am not, I will not release a song, or album if there is one note out of place, if the vocalist didn't sing one word right. or if the color of the album artwork is off by a shade, i will not release it till it meets my vision. Those who use these lazy tactics of technology should not be looked up to as great musicians, bands, in fact their musical standing, credibility or whatever you want to call it should drop by 10 each time they use lazy tactics, If you can not sing, hire a vocalist to sing the part. but using A.I. i am totally against it. They already trying to replace crappy pop artists with a AI Artist called Miquela. There needs to be a limit and no it should not be easy to where anybody especially the lazy and those with no talent and no true passion for music to record and release crap and pass it off as music. But that is my stone cold brutal absolute truth opinion and as the saying goes opinions is like A--Holes, everybody has one. so i will say i am simply against using A.I.
![Felix "Unfair" | [Stray Kids : SKZ-PLAYER]](http://i.ytimg.com/vi/Oswujxm2Ag0/mqdefault.jpg)
Bro, this is a great video. You really perfectly blended what is essentially a gear demo/product placement with a practical application and a lesson in how to present the best possible version of an idea regardless of technical limitations. I’m a crappy singer who communicates toplines for my “backing tracks for the song I forgot to write” with VSTs and such, and this totally nails the target audience (with my sample size of one 😂)
They need a fry setting or vocal tuned more to harsh for rock and metal type music
For now, probably just run the resulting track through a distortion filter. :)
You ACEd this video!
I'm sold!
I've done this to send to singers to sing for real. I'd say that's the best use for this kind of tool.
Edit: oh
hahaha yup
Ah man! You ripped into me in your song competition for using an AI voice. How the tables have turned!
i see it as he was saying, a tool for demos and what not. at the end of the day he'll want an actual singer to sing it. i always felt the competitions were more about final products.
That competition was so long ago that the ai voice probably sucked.
So he should. Dont use this shit on a final product FFS.
You weren't paying him. 💵💵💰💰
@@CreativeMindsAudio I love how literally no one reads the info about the contests before making comments on it. You have half a point, but no, it's not about a finished product. It's about a finished SONG. AI is obviously strongly discouraged because a song is about the performance, and AI can't "perform" per se. It requires absolutely no skill or artistic effort. But the production can be FAR from finished and still land as a valid submission. You can have a finished song and it be not even close to what you'd hear on a release, and AI definitely isn't something a person in their right mind would put into a finished song, except... MAYBE... Eleven Labs voice modding, but that's still a real singer, just with a drastic timbre and tonality shift.
This looks really good for people who still like to sing good but they want to sound great.
Ugh, it's actually affordable. I wonder how well you could use it to fix some of Suno/Udio's mistakes (if inpainting doesn't fix them), like extract the vocals replace them with this?
Really great demo video of this. Well done. Just to clarify a little off topic for your sake as well as other viewers...as Songwriter in Nashville of 16 years interacting with Publishers, writing and pitching songs and studying with numerous Hit writers...it's the general consensus that the vast majority of Female Artists actually PREFER Male vocals on songs getting pitched to them. Reba especially but most others as well. They often don't want a vocal that leads them too closely in artistic expression as they want to add that of their own. Just an FYI :)
That's pretty interesting to know. Even if it's way outside their range? Like they'd rather hear me falsetto it? Or I should sing it in a good key for my voice and they'll put it in whatever key?
@treyxaviermusic they just want it in whatever key as they and the producer will change the key to suit their voice
Trey stop breaking down my excuses godamnit!
Next thing you know, you'll put out a guide to fix my laziness.
Yes please do that. Maybe I’ll get myself to record my own instead of watching Trey make vocals about farting into a can 😂😂
If lazyness has always been a very serious problem in your life, to the point of turning your life into a mess, i strongly suggest that you take some time to study about ADHD (Attention Déficit Hyperactive Disorder).
Although it has the "Hyperactive" word in there, it's important to know that this disorder CAN destroy the person's ability to feel motivated or to sustain motivation, for this reason, a LOT of people with ADHD are HYPOactive instead of being hyperactive.
People with severe ADHD that suffers from severe lack of motivation are ALWAYS labeled as LAZY. They themselves grow up believing they are lazy and feel a lot of guilt and shame.
So, they are filling the gaps one by one
If you can't play instrument but can sing use this tool
If you can play guutar but not piano or drums use this
If you can't sing but play an instrument use that
If you can't compose or write a lyric or mix or master use this or that
Feeling me now
It's only a matter of time before it throws everything away including us as well 😮
I don’t think that’s a problem though. It’s good for people who use instruments as a means to an end.
There have been pedals on the market that have done that for a while now, like TC Helicon's Play Acoustic. Voice tune, pitch tune, octave change, gender change, accompany, harmonics, etc.
My favorite part is when you said the kid voice was 'scary' (KOwaii) vs 'cute' (KAwaii) lol. Solid tool though..gotta check this one out.
Is it possible to purchase an external voice and use it as a custom feature?
I know that A.I isn't sentient, for now at least, but I still think that we should respect it and appreciate it.
I have nothing against sponsored videos, when it is disclosed as you did. The problem in this case, I think, is that it prevents you from informing your viewers about the many options available to do this. I prefer Synth V, which is remarkably similar, but is not subscription model. Others who can sing reasonably well might prefer “voice changers” such as Audimee (there are others). Thanks for your video, anyway.
I used to try to use similar tools like SynthV and Vocaloid before I got better at singing. My biggest complaint with tools like these is that they almost always sing with so little power and attitude/personality, which works for a lot of genres but not quite so well with rock or metal oriented music. I'll still occasionally use it as a writing tool, but it can be hard to really conceptualize what the vocals would sound like with a proper belting singer sometimes
VocalSynths can sing with a lot of emotion, you just have to know how to do it and have the right voicebank. I feel UTAUloids are better about having emotion, but again, depends on your ability and the voicebank
So you draw a firm line between this and the generative AI being “cheating”…
So what if it’s being prompted using my original music and lyrics? I realize this video is a couple of months old already and that’s ages in the AI world. But I’ve been using Suno exclusively for vocal tracks to use this way. No, I can’t input the exact midi notes or the timing but that’s what makes it so impressive. It understands what to do with it even better than a real vocalist would. A certain number of prompts will give me ideas that don’t sound like what I want for my song and even if they sound pretty good, I delete them and move on until I get something that is true to my creative vision. Just like yourself, I would love to use these as demos to record with a real vocalist. But my aspirations as a musician aren’t focused around recording professionally or putting a band together. I just want to finish songs I started 20 years ago. And anyone who calls it “cheating” can just suck my balls 😊
People don't use a loom today. If we get what we like it's kind of fine.
i really like the concept so I took a look at their prices. If someone is a full-time songwriter who can't sing, this would be an excellent tool at a great price. I'm a little concerned that the price kinda puts it out of reach for the average "in my spare time" musician, though. That's about as much as some plugin suites for something you might only use a few times a year.
but now that it's out there, there's enough opensource ai tools available that I could see a FOSS version happening.
Hi.. Does the AI signer work as a creator of a vocal lines and melodies? Like if I have song without vocal parts, I'd would upload my song into the AI and it will create a vocal lines?
No, you have to write the part and it sings it
It's still got some odd sounding artefacts and unnatural decay on the ends of notes, making them sound clipped, but in 10 years this technology will probably be amazing.
To do extreme metal vocals, one almost has to be a cartoon voice actor, with various ranges and timbres.
If they can do Cradle of Filth style vocals with voices/clean singing/fry screams etc., it will be something.
It isn't just about the notes. It's about tweaking vocal weight and timbre.
Right now this seems really tedious to get right. Like tweaking velocity/timing settings on hi hats for a more human sound.
Two things:
1) In my experience, there is NO better way to give yourself a solid, intuitive sense of how to write vocal melodies than taking voice lessons. As in, within my first month and a half, my vocal melody writing became ten times better than it was before lessons.
2) BOY OH BOY, would this tool have been life-alteringly helpful in the days before I took voice lessons. Holy shit.
Yeah, I've had people send in songs that were written with AI voices but it sounds like they've never heard a person sing before, like they wrote th melody on guitar and it's something no person would ever sing
I can use this for my female background vocals and demos!
I am not sure if i understand. Can I replace just words in an already existing song with this tool?
Great tutorial and hilarious. I"m going to see if there's a metal type voice
Finally! I can make myself into a vocaloid.
This is how EVERYONE on the pop charts creates their vocals.
And that is AFTER someone else wrote the music, and someone else wrote the lyrics.
And someone change their faces, tits and asses....lolThe world is crazy dude
Yeah no
This is untrue lol
They want credit card details to start the free 14-day trial. They can go to hell.
😂😂😂
Im with you...101%
Since it converts to midi, is the quality of the microphone / recording out the window too? I assume that is the case similar to drum performances converted to midi are only looking for the hits and possibly velocity
It means that it doesn’t matter what microphone YOU’RE recording on, it’s always going to sound like the one THEY used to record the voice they trained the AI model on.
@@treyxaviermusic that's what I was hoping. As long as the AI can "hear". I think I'm sold on this
Trey, get ready for the massive amount of AI voices every friday now 🤣
I did it to myself
Yo man! Did you try SynthV before? I think SynthV can create a more realistic voice and it is a life time license. ACEstudio subscription is pretty expensive man😅
I’m ok with using AI as a brainstorming tool. Don’t use that in the final product though. Recreate it, even if it’s not 100% perfect.
im against most ai use but this seems fine to use as a last resort. im just some guy in my basement and if i wanted a female singer, i dont have the resources to get that in the first place. i think it depends on your situation. if you can do it for real, do it 100%, but if you cant, i dont think its that big a deal.
This is actually a pretty cool tool
all you have to do is mp3 out the midi voice and put it in a celeb ai voice thing, singer wise. i did a ghost song and just used
tobias Ai lmao
Not gonna lie, this software reminds me a lot about Vocaloid (using a piano roll and phonetics, amongst changing vibrato and other properties), but on strong steroids.
Even trying to go into this with healthy skepticism.... yeah my mind was blown in less than 5 minutes. WTF
what AI is that?
The future is wild
not in a good way.
Can it do low screams and pig squeals?
no it is a poser
@@treyxaviermusic Thank you for taking the time for that awesome answer!
If this is trained ethically, then it's awesome. If not, I just despise it as any other Gen AI.
All of the voices are trained on real people that they contracted with as far as I understand
Stan wasn't lying. He IS Lorde
I have a good singing voice but have issues singing without hearing backing vocals. Does anyone have any suggestions for getting better without them? I know this is more a guitar and songwriting channel. I think honestly its just practicing singing to instrumentals
Practice more
Backing vocals ad lib over the main vocals. A rapper with a hype person could have a ding in the head phones to note when to start and start again. Adding and then taking out the sound markers shouldn't be a problem for a sound engineer using multi-track.
Yeah, I know, I'm not taking it personally!
:40 - IMO this is a false dilemma. Even if the primary tool is generative AI (such as Suno, Udio), there is still tremendous power within the interface (intricate prompting, lyric input, stem exports, etc.), and this will only grow more in the future. There's also an aesthetic evaluation by the user that comes into play as outputs are culled.
Now, if ALL one is doing is typing in something like "catchy pop song" into the AI and accepting the output and sharing it, we're in total agreement there. Maybe I'm optimistic, but I think listeners will soon be actively listening for such work, and weeding out a lot of it.
Good channel, I should say.
Really enjoyed this video. I like the idea of being able to get female vocals from my voice. I might jay say dirty things to myself though 😂
It's fucking fantastic! It can not be that easy! Why am I a lazy human, not a some kind of super duper soulless machine)
I wish they turned it into a plugin
Trey was this targeted at me
lol
Insane
All in this "ACE promo videos" forget to say most important thing - ITS WORKING VERY SLOOOOOOOOOOOOOW....., not like in this promo videos.
This is actually pretty amazing. Great tool!
That final harmony was spicy and beautiful.
just click on the link guys
I remember when they came out with auto tune, and THAT seemed impossible. THIS is unreal. Also, I am old. Also also this video made me literally LMAO.
Can you convert this into an AI Celebrity voice like say Taylor Swift?
Seems like a great tool but not sure if I'd get much done as I'd waste hours punching in the darkest lyrics possible, including all the "RAWR"s, "BLEGH"s and "ARF ARF"s, but use the sweetest angel voice available, just for giggles.
I have done that, and got my song gonged by Trey because the vocals "sounded uncanny and AI generated"
No comparison in quality to vocoflex/synthesizer V. I don't see any professional use here.
Very Cool! Now if We could get something a little More Metal ala Rob Halford ,Bruce Dickinson for some good Power Metal, maybe King Diamond. or just go for Guttural death Metal stuff. still though. like I said. Very Cool!
Losing my job to AI? If that job is singing, then I hope so! Some nifty tools are in the works...
One problem with this is that some “creators” will undoubtedly release songs with these AI vocals on the finished product, possibly as the only top line. Your expressed use for this tech makes sense, but people will not adhere to it. Social disrespect for this choice will be unsufficient to stop AI voices / top lines from becoming normal.
What happens in 3-5 years when this technology’s update or successor provides an option to write melodies as well? AI will be used to “compose” every aspect of music that we consider intel property.
I love what you do Trey, but it’s irresponsible to promote such tech developments.
There's already loads of AI that does that kind of writing, and it's terrible, but this isn't that. Just because lazy people will abuse this and put out music with this as the topline doesn't mean it shouldn't be promoted as an incredibly useful tool. Lazy and uncreative idiots will always abuse any new tech.
@@treyxaviermusic I hope you’re right. Just to balance out my negative comment, I will admit that you are, IMO, the most inspiring musician and teacher around. I look up to your musical knowledge and opinions more than those of anyone else.
Hopefully in that case, that lame arse music gets the attention it deserves, i/e ZERO. GENERATIVE A.I means you get fuckin zero respect and thats the way it should be, forever.
@@treyxaviermusic Actually I've just recently heard a few AI metal songs and they were VERY good! 😮😭 I mean, there was some hiccups vocally, but the riffs and melodies were there! And the reality is there's just SO much (real) music out there these days that it's hard to stand out from the sea of genericness. So, when u DO have a good riff AND good melodies u DO stand out! Thus these songs were some of the best I've heard in awhile and I hate them for it! 😫
It's always a complex subject. A tool Vs an ai that makes things to you. Respect man, but, if u can't sing and the aí sing to u, is the second type of ai. U know how a good singer sings, but u can't. A person can know how a good drummer plays, but can't do it. I think in the future we will have more "producers" than musicians. 😁. Complex subject.
Hey ❤❤
A subscription service. Pass.
This is definitely NOT for me. My attitude is that if you can't sing. Just don't. Find somebody who really can.
That's the point of this - you can use it to write something for that somebody to sing. Plenty of great singers can't write for shit
Good point. Ai tools are exponentially getting better by day. Very exciting. Can’t wait for a AI co-producer / mixing engineer to make productions with.
“More bongo’s jives and a new vocalist… yes please”
He suggested using it to pitch a song as a reference track. Not to put it out as an actual song with the AI lol
I think people should learn to sing as soon as they begin to play guitar.
🤗😝
This melody is so good!!!
The one at the very end of the video was the best take I think 🤣🤣🤣
It will be a problem for foreign languages
Hear and look the Video from the Recordingsession from "We Are the World, we are the Children" produced from Quincy Johnes, performed from
•Lionel Richie
• Stevie Wonder
• Paul Simon
• Kenny Rogers
• James Ingram
• Tina Turner
• Billy Joel
• Michael Jackson
• Diana Ross
• Dionne Warwick
• Willie Nelson
• Al Jarreau
• Bruce Springsteen
• Kenny Loggins
• Steve Perry
• Daryl Hall
• Huey Lewis
• Cyndi Lauper
• Kim Carnes
• Bob Dylan
• Ray Charles
and many others in the Choir. After you have hear this, you will know what it mean to sing Musik with all your Soul, all your Emotions, all your Energy and how it sounds,
if realy skilled and trained Singers, real Professionals who are real Artists, performing real Music !!!!!!!! All this AI faked "singing" is crap for the trash !!!!!!!!
AND THE LAST WHAT WE NEED IS A OTHER SUBSCRIPTION !!!!!!!!! SUBSCRIPTIONS ARE FOR PEOPLES WHO HAVE A MONEY SHITING DONKEY AT HOME...
...BUT MOST PEOPLE HAVE NOT SUCH A DONKEY AT HOME...
There are plenty of these tools. I just wonder why anyone would need an AI singer. AI doesn’t have an artist persona. You can't perform live with AI singing, and you can’t create any content with AI singing other than music without real people behind it. Who would be interested in that?
He's obviously being paid to say how good this is. The tech does show great promise, but the free version isn't very good, and doesn't do all he says it does.
Sending farts to her dad??? 😂😹
Holy shit is this actually real?
It isn't that I can't sing...it's that I have a weird and completely illogical phobia of singing.
Sounds like this might be the thing for you!
@@treyxaviermusic Looks like it just might be...I can just talk and it makes it happen. Wish it wasn't damn subscription though!!!
I want to hear your terrible falsetto impression of what she's supposed to sing.
Irritated by the start, I later on felt like saying thank you for making me - at least feel like about to - understand what's so relevant about AI singing: if you find a melody you can have the best or the worst to sing it. I would like to add - having seen all - that the presentation, the performance of any song is decisive. Like, quote, Tina Turner, when asked to sing a song: "Yes, I can - do it", i.e. make a melody a song. In fact, same thing, said from the other side. What you emphasize right from the start, to me, seems that from now on any tune or combination of tones can be "turned out" a song, if it fits the right voice.
Also, singing - and only singers would disagree, at all - is clearly defined as a skill just like any other.
Question: don't you need great voices for the input, from the start? The example you give at the beginning of your video looks like an argument.
I can’t stand the sound of heavily computerized voices. There’s no grit or emotion. Boring.
I so totally agree to this! How in the name of our Lord and saviour were people in the past able to convey other people how to sing without AI! Man, can you imagine the songs we could have had! Forget all the hits of the past, we need to start over! With AI! I'm not taking one subscription, I'm taking three! My music will be so good and singers will love and adore me now!
DON'T KID YOURSELF PAL, tell that story to a professional singer,instead of justifying what you are doing. AI is what it is. It's acceptable for you to replace a singer because You are not a singer. If you see you wouldn't have that flippant attitude. When music production is replaced, come back and let's see what your attitude is about “musical integrity". AI is great until You loose your clients.
Can you make your own voice sound better instead of making you sound like someone else?
S.A.A.S S.U.C.K.S. Synthesizer V is much better & can be purchased outright.
NO INDONESIAN !!!
I mean obviously this is a great tool, but you just know some people are going to use it to cheat and it destroys my faith in music not to know whether I'm listening to AI or not. That's why these things suck.
That's the large-scale problem with AI though, and I very much agree with you on that. People always abuse technology, it's why we can't have nice things. But we can't let those assholes ruin everything
@@treyxaviermusic Narrator: The assholes ruined everything
i can sing, but i don't like my voice. so i sing and record it and then swap the voices, i don't think this is cheating. ok, the people who can't actually sing would cheat then, but it would be a nice tool to compose, so its still human working on tracks.
It won't be a problem.
The greatest artists of the future will use these tools too, and we'll think: "Daaamn, i have access to the same tools this artist has but i can't create amazing songs like him/her."
In reality it will be good for all of us... because this technology won't make great artists worse, in a matter of fact it has the potential to help them make more and better songs.
At the same time it will benefit a LOT of people that struggles too much to finish the songs they started creating.
At the end of the day we'll discover a lot more songs that we like (and a lot more that we hate, but these we can ignore after we tried it)
@@brunomartinello1114 I disagree. I'm a songwriter and musician, and I can't think of anything I want less than to have something write part of my music for me. If any great-sounding music arises from AI, it will be soulless by it's very definition.
Yeah, OK so if I have a beautiful song sung by a male singer, you telling me a female vocalist will have no idea how to sing it unless she has a female on the demo come on, dude that’s ridiculous
No that's not what I'm saying - it's about presenting it in the best possible way to sell them on it. Obviously she can hear my terrible falsetto version and sing the same thing but better - but will she WANT to?
Using AI to fix vocals isn’t cheating??? Have you learned nothing from Milli Vanilli?
Using AI vocals as a stand-in for songwriting demos isn't cheating
@@treyxaviermusic okay, fair enough, I’ll give you that. 🤘
@@treyxaviermusic I did the thing man, I’m sorry. I reacted before watching the whole video.
Anyway, I do agree that AI can be a great tool for figuring out concepts, as long as it isn’t used AS your final product. Unfortunately, too many people will use it as such. Keep rockin’ bud! 🤘
creators will save the planet
I might have been willing to try this until i saw it's subscription only. Never going to sub.
But when the producer finds out how much the artist will cost that producer will just keep the AI version. 😢
Producers don't pay the artists, that's not how it works
@@treyxaviermusic So then enlighten us..
Dont bother pitching it to a female artist just stick with the ai woman you used.
Nitpick much. It’s a demonstration. You do you.
@@johndo9648 I'm just saying if you go to the trouble to create it why not just keep it? 😁
I would caution all on use of AI tools for vocals AND instrumentals, like this and Vocoflex, to use them only for prototyping if you care about reputation as a legit human artist. Vocoflex imprints their output stream with an indelible watermark; others, if not all will follow. If anyone declares that your music is AI and then checks for the watermark, they could then claim "you're nothing more than an AI artist! 😡", even if you used the AI tool as... only a tool... as you would any other production tool. However, in the case of AI tools, people will be quick to slander you as an "AI artist" and get others to dismiss you on that account. Don't get me wrong... I LOVE AI. In fact, I work in the field and don't see any way forward but to embrace AI. But that's because I am pro AI. Most people won't be, especially as job loss due to AI increases to a fever pitch. Be careful of how and when you use these tools, especially when publishing. Just an early warning.
But Trey, we're not here because we create trashy generic pop.
What you make with the tools is up to you
What an awful melody makes it hard to like
No need to illustrate how juvenile one can be. Better to keep some degree of decency
yeah, wouldn't want to have any fun, much better to be a fucking wet blanket
No need to illustrate how lame one can be. Better to keep your lectures to yourself, instead of being a turdwipe.
first off i have issues with the crap you and other "SOCIAL MEDIA INFLUENCER MUSICIANS" spew the whole B.S. of constantly put new stuff out don't worry that it's not perfect. Me i am one of those true musicians who believe you do not release stuff till it's absolute perfect and meets your vision 100 percent i started playing guitar and writing songs in 1990 and i have over 3000 songs, or should i say song ideas since they are partial lyrics or just titles, since my playing skills are not up to meet how i want the songs to be. I will not take the lazy way out of using plug ins, studio musicians, drum loops, click, kick tracks, backing music tracks, or cut/copy paste instead of the old school way of double tracking, layering and recording parts buy actually playing the parts over and over again. ( wont go into why that is better and you get more out of that than using the lazy cut/copy and paste. also Modern day musicians/artists/bands have become too lazy and dependent on technology. MY motto is never put trust or faith in technology cause it will fail and or betray you when you need it the most. As for AI for singing. I can not sing my way out of a wet paper bag. My voice when i sing sounds like music played on a cassette that the cassette player is slowly dying and the music is slowing down with the vocals. That is how my voice sounds.I have a vision of what my music is supposed to sound like and I will not compromise it even if it takes me till i am 99 years old to release my first song/album. Those who use these lazy ways of doing Music whether it is click, kick tracks, drum loops, backing tracks for the music, Singing, Musical instrument parts (I have no problem with backing tracks for sound effects like crowd noise, gunfire, lightning, thunder, or narrative speaking like the prayer in Metallica's enter sandman or at the beginning of Megadeth's Captive honour track from countdown to the extinction) Then there is the lazy vocalist vocals, which is known as cookie monster vocals, or i call it either angry dog barking vocals or more accurate the lazy vocalist vocals, you know them as growl vocals. then there is the lazy using plug ins, computers, drum loops, instead of real instruments, real amps, real speakers, real speaker cabinets. I am tired of the whiny safe space/cry closet woke , entitlement snowflakes whining about it costs too much to travel with real amps, cabinets, rack effects, blah-blah -blah. Guess what that is a part of being a musician DEAL WITH IT. I will never to the falling in reverse Ronnie Radke thing. Me and my future bands will have real amps, real cabinets and real speakers, racks and pedalboards, and since my future band will have 2 drummers and 3 guitarists the drumkits will not be these what i call kiddie drum kits, They will be full oll out kits, like Neil peart and Terry Bozzio, etc. Also will not allow producers no matter how famous they are, do what i call the Dann Huff Move (you know without band, artists knowledge or consent, either they or a studio musicians they hired changes and or adds parts and or remove parts behind the artists/bands backs, as Dann Huff did with Megadeth) I will not tolerate that from any producer no matter how famous, if they did that i and i found out they would be fired on livestream so the world could see it and they would not get paid for their services cause they would have broken the agreement between producer and client. Right now todays musicians, artists have become too lazy and dependent on technology. Dave mustaine was 1000 percent correct when he said something to the effect that due to technology there is a lot talentless people recording and releasing crap and passing it off as music today. Guess what when you flood the market with crap, guess what happens, consumers start to buy less or stop buying all together. That is why the album sells, cd sells are way down. also bands, artists bought into the lie from record labels, social media clowns or whoever and that if you don't release new content, material every 3 to 9 months, then people will forget about you and move on to the next flavor of the moment. I can easily prove that to be b.S. Guns n roses Chinese Democracy album took 8 years and multiple lineup changes and when it was releases there still was tons of fans buying in and going to see them in concert. Putting crap out every 3 to 9 months hurts the quality of that material. You need to spend more time on the music. Otherwise you get filler songs. when people did buy albums, and artists, bands spent like a couple years to release music. You make get 1-3 and very rarely 4 filler songs out of the standard 10/11 track album. (yes i consider 10 or 11 tracks is standard for albums) today you rarely get 4 good songs and even rarer is getting hits. most of the songs on albums today are filler songs. they are as crappy as these left wing reboots of classic movies and tv shows. Me my sole passion is music and i will never compromise my vison for my music even if it means i will be in my 90s before releasing my first song or albums. other musicians, bands, artists, and "Social Media Influcencer Musicians" who are a lower class of musicians. May have no problem with releasing crap, constantly till they get to the jems. I am not, I will not release a song, or album if there is one note out of place, if the vocalist didn't sing one word right. or if the color of the album artwork is off by a shade, i will not release it till it meets my vision. Those who use these lazy tactics of technology should not be looked up to as great musicians, bands, in fact their musical standing, credibility or whatever you want to call it should drop by 10 each time they use lazy tactics, If you can not sing, hire a vocalist to sing the part. but using A.I. i am totally against it. They already trying to replace crappy pop artists with a AI Artist called Miquela. There needs to be a limit and no it should not be easy to where anybody especially the lazy and those with no talent and no true passion for music to record and release crap and pass it off as music. But that is my stone cold brutal absolute truth opinion and as the saying goes opinions is like A--Holes, everybody has one. so i will say i am simply against using A.I.
k
@@treyxaviermusicThis dude is not into “the brevity thing”…
Feels like an AI 🤖 Wrote that 😂😂
nobody stops you from doing that.
Your last sentence could’ve been your entire post…