Elton John - Can You Feel The Love Tonight (AI generated - 70's voice)

[@PauloHirth]

Preservar a voz de cantores fenomenais como Elton John é a mais encantadora maravilha da inteligência artificial!

[@joneslee2425]

동의합니다. 위대한 가수들은 이제 ai 를 통해 영생을 얻을 것입니다😊

[@Kreat-ik5uc]

If someone told me that Elton actually recorded the demo of this song in the 70s, I shall believe that word. I wonder Elton would sob if he heard this voice...

[@hockley91]

I'm still blown away at the clarity of it. I took all of these isolated vocals and made on big .wav file for the AI to work with. Just incredible.

[@dogpd3]

@@hockley91I wonder if Elton will use AI technology to do a new album and use his “younger” voice with the new songs. The possibilities are endless

[@Riri-oj1zs]

@@dogpd3 I do believe people would love to hear his young voice again using his new compositions. This could be the norm in the future.

[@milamarinett8046]

@@Riri-oj1zs Of course, artificial intelligence can do a lot and sometimes it's useful. But I am against technology replacing real people. Anyway, the real Elton would have sung it better, with emotion and more soulfully. And with the help of such programs, he just sings a set of sounds.I like his voice both in his youth and now, but his voice of the 90s is better. It is stronger, cleaner and more powerful.

[@Riri-oj1zs]

Elton's voice in his prime was just impeccable!!!

[@jonkeiser6092]

I bet Elton John would be proud and say, there’s no way I can make my voice that high anymore.

[@heyitsroddy]

Literal tears. That voice.

[@hockley91]

It turned so much better than I thought

[@BabyLambCreations]

The moment when the greatest and cleanest AI cover is from literally the same person who sang the original. Still crazy to hear the difference between Elton's voice in the 70s and Elton's voice since the 90s. Regardless, really great putting this together!

[@hockley91]

Thank you! It was a lot of fun to do this!

[@andrewwagner7992]

Amazing, what a time to be alive.

[@znrctrnn]

Great to sing along with since I’m programmed for the 70’s.

[@andrepaz9219]

Simply a masterpiece!

[@juanpabloeg6222]

Please, I'm Gonna Love Me Again with young Elton 🙏

[@j65055]

Esto de la IA es tan mágico como aterrador. Me encantaría poder escuchar las canciones que Elton ha grabado desde 2001 con su voz de los años setenta u ochenta.

[@Its_King_7]

I've Listened To This So Much, Thank You For Making This, Can You Do Cold Heart Maybe?

[@harshmahatme9005]

Please do a 70s AI version of the song from his movie 'I'm Gonna Love Me Again'

[@PeteBalikov]

AMAZING! Can you tell me how exactly you made it like this? I would like to make a cover of my mom's favorite songs, she loves young Elton.

[@vladroit]

that's beautiful ❤

[@Гаксим_Малкин]

Отлично получилось

[@theperson_in_thesuit]

Good

[@garridotito7570]

El viaje en el tiempo si existe

[@meldreamsdocometrue]

In memory of Frank Wells- a former Disney President

[@joneslee2425]

20대 엘튼존 선생의 목소리를 들려주셔서 감사합니다😂

[@HardKickinGroupGermany]

Yes! Not just a dry vocal. 😲

[@hockley91]

I added reverb to it.

[@tornshadows]

Amazing and so clear audio, yes isolating the vocals is the trick most of the times :) What program did you use for the process? Good work :)

[@hockley91]

Well, it was a lot of tests. Ultimately I found isolated vocal tracks of Elton John on KZbin and other places. You'll need to google for them. I found lot's of isolated vocal tracks. The best isolated tracks were from Elton that I have. They are clear and have no effects. Completely dry. I went through each file and chopped it up and kept only the good stuff. There was still audio bleed from the headphones he was wearing. I used RipX to separate the background from the vocals. I reprocessed the vocals and they sounded pretty good after that. This is the KZbin video I watched on how to take an isolated vocal and train the AI to emulate it over another vocal track: kzbin.info/www/bejne/noTbhX95h8Rsa7s I also watched this video to understand the errors I was getting: kzbin.info/www/bejne/goHXXnuvq8SCqpI After the AI file is created (which takes the longest time), I then just apply that file over another isolated vocal track. I experimented with other musicians. I even experimented with my voice on some live vocal tracks with the band I play with. It was amazing to hear Elton's voice over my voice with my piano playing. It helps to emulate the way they sing and it worked out amazingly. Here's me doing an impromptu performance of Tiny Dancer and Rocket Man from last year. Elton is singing over my vocals! Awesome! I messed up a lot, but I was doing it on the fly and didn't have the lyrics memorized: www.dropbox.com/scl/fi/sw73d13dkommzoojqntl2/GP22-Elton-John-Encore.mp3?rlkey=ub2qoyfz32x5yonih7z9bq9tk&dl=0 Here's the recording in my own voice. I tried to emulate his singing style as much as I could: www.dropbox.com/s/4ebyqlqdke9mlxa/2-13%20-%20Elton%20John%20Encore.mp3?dl=0 Anyway, that's pretty much it! It's fascinating and shows where this is all headed. We can recreate vocals, deepfakes, etc. There will be a full database in the future of a person. Lot's of data (text, images, video, etc.). Can we actually recreate a person in the near future? Interesting to think about.

[@jaggass]

@@hockley91 Elton always recorded the vocals last so there was less bleed from other instruments.

[@psychodied]

​@@hockley91incredible and amazing.

[@King_Colombia_Inc]

If The Lion King was made in 1975.

[@st0n3p0ny]

Wasn't this Elton in the first place?

[@hockley91]

@@st0n3p0ny it was his original vocal performance, which the software replaced with his younger voice that I isolated from various vocal only recordings from the 70’s.

[@samf_eater]

Say pls, where did you do AI generated track?