Live data from Hacker News

Lyrebird – An API to copy the voice of anyone

lyrebird.ai

41–50 of 311 posts

Re: Lyrebird – An API to copy the voice of anyone

#42
post #6

This is pretty cool (although, I have no idea what other technologies exist for this kind of thing), but it's definitely not convincing enough to a human listener. This sounds like it might be convincing enough for some programs like "Hey, Siri" but it's not gonna convince your mom. You can listen to the samples on the page linked here and you can immediately tell that Obama and Trump don't sound quite human.

Text to speech is still pretty distinguishable as not-human, and that seems like an easier problem (only has to work for one specific voice, not an arbitrary voice). So just on the basis alone I wonder if this isn't still a ways out

Interesting thought: is it easier or more difficult to make a synthetic voice undistinguishable from a human one, compared to producing speech copying a real voice?

To me, at least, the voices I heard in Lyrebird's demo actually sounded more 'real' than Microsoft Sam for example.

Re: Lyrebird – An API to copy the voice of anyone

#44
post #19

It's there any copyright protections for a person's voice? If not, David Attenborough and Morgan Freeman will be lead voice actors in my next game project

Copyright was designed to protect artistic expression. A song can be protected. A performance can be protected. A playing style cannot.

Re: Lyrebird – An API to copy the voice of anyone

#45
While all of these vec2speech type models are impressive, I get the feeling that most of the comments didn't listen to any of the samples. It's still distinctly robotic sounding, probably has quite a bit of garbage output that needs to be filtered manually (as many of these nets often have) and is a far cry from fooling a human.

Re: Lyrebird – An API to copy the voice of anyone

#46
post #20
post #19

It's there any copyright protections for a person's voice? If not, David Attenborough and Morgan Freeman will be lead voice actors in my next game project

AFAIK not copyright per se in the traditional sense, but there is a "likeness right".

This. I would assume copying a voice pattern is treated no different from copying an appearance. If a 3D model is a very close approximation to a particular person's face and you use it in a game intentionally to benefit from that person's likeness, you can run into legal issues. And of course you really open yourself up for a lawsuit if you then attach the real person's name to it.

EDIT: See the NPR Planet Money podcast linked elsewhere in the comments. Apparently there are fairly specific laws to protect someone's likeness -- including their voice -- thanks to the entertainment industry and Frank Sinatra.

Re: Lyrebird – An API to copy the voice of anyone

#47
post #11

I was wondering when CG Sir David Attenborough would get here and start narrating my day to day.

Actually, when the sad day comes and he passes away I'd be very comforted to hear his voice on new nature documentaries. I just can't watch them unless he's narrating.

Re: Lyrebird – An API to copy the voice of anyone

#49

While all of these vec2speech type models are impressive, I get the feeling that most of the comments didn't listen to any of the samples. It's still distinctly robotic sounding, probably has quite a bit of garbage output that needs to be filtered manually (as many of these nets often have) and is a far cry from fooling a human.

Sure, it's not clean and crisp but if you ignore the distortions and the somewhat arbitrary intonation it's pretty close to the real thing. And I would argue that it's already good enough to fool grandma for a phone scam (but then again scammers currently get away with practically no resemblance at all).

Re: Lyrebird – An API to copy the voice of anyone

#50
post #19

It's there any copyright protections for a person's voice? If not, David Attenborough and Morgan Freeman will be lead voice actors in my next game project

The only possible voice I can think of that I would be certain it could have this kind of protection right now is actually Majel Barret-Roddenberry. She apparently did a lot of voice recording specifically to let this kind of thing happen. What I couldn't fathom is what would happen if something like this was used to mimic someone else's voice that hadn't agreed to such a thing.

http://www.avclub.com/article/late-majel-barrett-might-still...

Post reply on HN