When I hear mixes that sound this good, I always think of how far we've come since the first vocal synths 🥺
@illusioncovers
2 жыл бұрын
thank you! i feel it. i have been covering since 2015 and even my own improvement through the years is fun to see along the way vocal synths as a whole have changed over the years...
@wildandbrey
2 жыл бұрын
@@illusioncovers I've listened to your stuff for a couple years now and you can definitely hear the improvement!
@LetzyPerez
Жыл бұрын
This is absolutely beautiful! She sounds just like an actual person!
@rmsv
Жыл бұрын
"She" is an actual person. She is Emma Rowley, a broadway singer/songwriter, whose voice was used to create this model.
@rubinchavarria7173
2 жыл бұрын
She’s borderline uncanny . Let her out of the basement 😂
@3dfantasymarkkun
2 жыл бұрын
My thoughts exactly!!
@English_Lessons_Pre-Int_Interm
Жыл бұрын
did you mean, let her play in adult films?
@3dfantasymarkkun
2 жыл бұрын
This is absolutely insane! The emotion, the nuance in tone... GREAT JOB! 💕💕💕
@silmist11
2 жыл бұрын
this is CRAZY GOOD theres so many wonderful little details it really sells the feeling of the song!
@alandavid4370
Жыл бұрын
Solaria is amazing goood 😍
@addd21323
8 ай бұрын
This is so insane lmao. Incredible.
@dannydancer1043
2 жыл бұрын
So beautiful 😭😭💖💖
@heath-7567
Жыл бұрын
A game changer in relation to our 32 song multi media project that just needs vocalising..🎤🎙
@Lunaar
Жыл бұрын
I'm speechless, i can't imagine the world 5 years from now, the IA will be even more powerful, woaaaaaaw
@GhostGuitars
Жыл бұрын
Amazing that things have come this far. One thing I would say that stands out as synthetic though, is it sounds like 'she' is singing in the same 'region' of the voice. I cant hear a difference between head/mix/chest
@illusioncovers
Жыл бұрын
this is definitely something would be able to acheive better now actually- theres been an update to allow better mixing of the vocal modes. additional i didnt use tone shift on this either which would have helped! (some voices are more clear at having a different though but they do admittedly specialize in specific areas. but it can be acheived with the right parameters imo).
@LeeBlaske
Жыл бұрын
Great job. This software is pretty amazing, but you have to know what you're doing to get results this good, and you definitely know what you're doing. I like how you left it exposed in a fairly simple arrangement.
@PhilReyGibbons
Жыл бұрын
I don't know what's best, the software or the skills you have to make it sound like this.. wow ! (nice mix and piano overall too). May I ask how long it took you to get there with the vocal part ?
@illusioncovers
Жыл бұрын
hi! thanks for the comment. this file was not tuned by me, in the sense that i downloaded one that one had already done all of the general work for. but i have tuned files in a very similar method, as basically this cover uses a tool called Praat, which allowed me (and the person who tuned this file) to export the pitch bends of the original vocalist from an acapella. there is a lot of cleaning up that needs to be done, like smoothing out the pitch bends and fixing timing, but the method is technically faster thing going entirely from scratch. In the end a cover like this would take me a few hours. but i have been using it for many years and work fast in general.
@PhilReyGibbons
Жыл бұрын
@@illusioncovers Thanks for the very detailed answer. Appreciated. So what you're saying is that you extract a sort of automation file (with portamento/pitch bends) from the original vocal acapella part with a thrid party software (Praat) and then apply this file to the vocal part you programmed with SynthV is that it ? It seems to me that in this case the real AI thing is done by Praat here ? Just to be sure to understand ;)
@illusioncovers
Жыл бұрын
@@PhilReyGibbons yes, thats basically it! there is a channel called HPsynthV that goes into detail about this plug in as well, as he made the plug in that converts this data to Synth V. Praat is typically used for text to speech actually, and just gets that pitch data and also time out the places where words stop and end (this is a very tedious, hand done process lol). while it certainly gives the sense of realism to the voice to mimic the pitch bends there is no real AI involved in this pitch converting process, Synth V's vocals still has the learned knowledge from AI however and puts a lot of the work into the render of the voice alone, with its own natural enunciation, and requires work on my end too as I can do things like change the length of every individual consonant and vowel. SO praat doesnt use any AI stuff here, just conerting raw pitch info (which can be very unclean in many cases, since i use an AI tool to seperate source vocals and instrumental)
@illusioncovers
Жыл бұрын
i have something I decided to do after explaining the Praat stuff, that may be useful to you in seeing the difference between the outcome of copying pitch data from an original vocalist, VS letting Synth V do it for you! kzitem.info/news/bejne/y32m1WqVcpFniGU in case you were curious too, I also showcase the raw outcome from Praat, to show how...not very good it can be haha.
@ElliottKayneMusic
Жыл бұрын
Incredible job on this one it's amazing 😎👍
@rimiulnamo2059
Жыл бұрын
brother, thanks for showing the parameter i learned a lot with them
@efexzium
Жыл бұрын
who needs a mic with this lol
@artessence2024
Жыл бұрын
cool!!!
@nnoiz
Жыл бұрын
great !! ^^
@TheRickymak
Жыл бұрын
I just discovered this vst, and I’m curious about how long you took also. Is it a steep learning curve? Nonetheless, this is an amazing cover. Thank you for sharing.
@illusioncovers
Жыл бұрын
i responded similar to someone else so i will copy paste + add a bit more: hi! thanks for the comment. this file was not tuned by me, in the sense that i downloaded one that one had already done all of the general work for. but i have tuned files in a very similar method, as basically this cover uses a tool called Praat, which allowed me (and the person who tuned this file) to export the pitch bends of the original vocalist from an acapella. there is a lot of cleaning up that needs to be done, like smoothing out the pitch bends and fixing timing, but the method is technically faster thing going entirely from scratch. In the end a cover like this would take me a few hours. but i have been using it for many years and work fast in general. the learning curve is not too bad imo. Synthesizer V has MANY tools to help, like a sort of auto-pitch bend tool that every AI voice has, where it will generate pitch bends based off of the vocals' own style! it is super useful, and a great point to jump off of.
@andrewworld3681
Жыл бұрын
wow, you mus do tutorial how you edited all those nuances :)
@kichihoney
Жыл бұрын
Stand by me please
@English_Lessons_Pre-Int_Interm
Жыл бұрын
Can this TTS read web pages? Is there any browser which can use it?
Пікірлер: 34