Subscribe
Sign in
VALL-E — AI that can mimic a person's voice with just 3 second sample
VALL-E emerges in-context learning capabilities and can be used to synthesize high-quality personalized speech with only a 3-second sample. VALL-E synthetically preserves speaker's emotion and acoustic environment.
17
Replies
Chris Messina
Top Hunter
Hunter
Now anyone can have James Earl Jones (or anyone else for that matter!) voice their radio ads.
Also relevant given Apple Books's just-announced Digital Narration audiobooks.
Share
@mefranco it's not publicly available to test yet, but if you follow the links, you'll see a bunch of audio file tests that show what it's capable of.
Wow! Only a 3 second sample.
I'd rather see some examples in the images rather than how it works though
Congrats on your launch
@chrismessina Yep. I was mainly referring to the PH post, not the website.
My bad, should have specified.
Congrats on the Launching. We need more tools like these to enhance our creativity. Keep going, guys!
