Live data from Hacker News

AI Clones Your Voice After Listening for 5 Seconds (2018)

google.github.io

171–180 of 338 posts

Re: AI Clones Your Voice After Listening for 5 Seconds (2018)

#171
post #99

If acting was ever a good career choice, it isn't now. I am becoming convinced that actors will be replaced wholesale.

No, actors are still going to act. But their voices will be just an input feature to the speech engine.

Meaning that this democratizes acting. Celebrity actors won't be required.

I think this is a good thing for the field and gives a whole lot more people access to the opportunity they strive so hard to achieve.

Re: AI Clones Your Voice After Listening for 5 Seconds (2018)

#172

Do papers like this have code to play with anywhere?

from two minute papers video description: > An unofficial implementation of this paper is available here. Note that this was not made by the authors of the original paper and may contain deviations from the described technique - please judge its results accordingly! https://github.com/CorentinJ/Real-Time-Voice-Cloning

Re: AI Clones Your Voice After Listening for 5 Seconds (2018)

#173
post #42

Lyrebird has had similar technology for a few years now. https://www.descript.com/lyrebird-ai

Worth noting that Lyrebird is very rough -- at least last I tried -- and produced extremely robotic sounding (though recognizable) audio.

This method has much clearer audio, but seems to lack generality / TTS capability.

Re: AI Clones Your Voice After Listening for 5 Seconds (2018)

#175

Earlier quoted context omitted.

Philip K Dick wrote about people going to commune with artificial personality constructs of their deceased loved ones. Unfortunately, it's been a long time since I read it, so I don't remember which book it was in. Maybe someone who's read him more recently can remember. Update: Apparently, lots of other people wrote about this too, but PKD wrote about this before any of the ones mentioned so far, as he wrote about t…

Ubik? Though that is not about artificial personality constructs, it's about communicating with loved ones in half-dead states.

Yes, Ubik has half-life states.

But I'm thinking of a different PKD book where there were actual artificial personality constructs instead.

Re: AI Clones Your Voice After Listening for 5 Seconds (2018)

#176
post #100

At first glance there seems to be more more malicious uses than good ones with voice. Yes, hearing someone dear to me say things he/she never said maybe comforting. Anything else? Maybe some movies with the deceased actor's voice? But what if someone who wants to hurt me sends me files (or phone calls) from the deceased person saying horrible things like: - "I am still alive but left as I was tired of you" - "oh Jan,…

Maybe a little less insidious than your examples, my first thought was being able to generate voice actor lines for videogames. Maybe not main characters, but background NPCs and the such. Might make the VA union a little nervous going foward though!

Re: AI Clones Your Voice After Listening for 5 Seconds (2018)

#177
post #24

Anybody else notice how that Scottish male reference voice sounds considerably more English in the synthesized versions.

Yeah, maybe it's just having a more sensitive ear to the Scottish accent but that to me was the furthest from the reference by far.

I heard that too; being a Brit (and working in languages) probably helps. It did pick it up occasionally though, which gives hope that increased sampling and training could fix the slight miss there.

It was that and the Swedish-accented English ("Sentence in Different Voices" section, middle recording) made it struggle. No traces of the Scandi-lilt were left in the synth version.

Final note would be the French speaker at the bottom of the page seems to be English first language, despite having very good spoken French. Not quite as pure a test of that last part as I'd have liked, despite the ability for the speaker to perhaps read the synthesized version in English back in English. That could be fun.

Re: AI Clones Your Voice After Listening for 5 Seconds (2018)

#178

Now we can have audiobooks read by anyone we like! They can direct us to our destination! They can speak at our funeral, being long dead themselves (as long as there is sufficient training material recorded). The future is awesome.

I did something like this before my grandmother passed. She was a teacher and loved reading books to kids. I recorded her reading Dr Seuss and the Giving Tree to my cousins so I could give my future children a glimpse of that wonderful woman.

It seems that we aren’t far from being able to take those recordings and spin it into a reading of anything. Fascinating. It’s kind of scary though. Grandma’s voice can read anything. Anything.

Re: AI Clones Your Voice After Listening for 5 Seconds (2018)

#179
post #65

Earlier quoted context omitted.

When I interviewed Ray Kurzweil we talked about the obvious-in-hindsight insight that his life’s work was essentially trying to build an AI to bring his father back to life.

Except no matter how good it is, it will only reinforce that it’s not real and that he’s gone. Perhaps that’s the therapy he needs to move on? Note: this is different from listening to recordings from the actual person. Having loved ones die is one of life’s universal terrible qualities.

I think a lot of people's passions are driven by a hole in their heart, that they hope their work will help fill somehow. I suspect that no small amount of the enthusiasm for XR is due to a deep and abiding desire to be someone else, somewhere else, among the people developing or early-adopting for it. Of course, it doesn't have to be so high-tech; much non-profit or social work is prompted personal experience with the presence, or lack thereof, of the service being rendered.

In the end, I don't know if any of that works. But what's being subscribed doesn't seem too far outside the norm. Deprivation often leads to desperation for even a taste, however imperfect it may be.

Re: AI Clones Your Voice After Listening for 5 Seconds (2018)

#180
post #138

Earlier quoted context omitted.

It means you can make someone talk, non-stop, for an indefinite amount of time. Now I want to make an art piece that's just a valley girl droning on, and on, and on, and on about believable and obnoxious life experiences. The stores they go to, how they feel about certain colors, what "too spicy" is. It just never stops.

This sounds like a brilliant idea, and I'm almost wanting to build it myself. If we can tack on a visual head and sync up the lips to the words, it'd be amazing and would likely go viral.

ShadyWillowCreek@gmail.com if you're actually interested in collaborating on this.
Post reply on HN