Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

15.ai

15.ai

Post reply on HN
Print view View original

31–40 of 192 posts

Re: 15.ai

#31

Post by paulsutter »

marcofatica wrote: post #4
armchairhacker wrote: post #2

This is really cool. It's a text-to-speech and the gist seems to be that they synthesize it from only a little audio. The results are clearly synthetic and need work. However what's cool is that there are a ton of characters (from popular shows and video games) and there are useful statistics like inferred emotion (which is also in the output). Honestly it's a big problem how a lot of AIs are like "black boxes" where…

> But seriously: how did you get the domain `15.ai`? it's an MIT project so I'm sure that was a factor

.ai domains cost a couple hundred bucks a year so domains are very available / not widely used by domain squatters (Its the country domain for the island of Anguilla, pop 15,000)

Re: 15.ai

#33

Post by nope96 »

Oh god, 50 shades of SpongePants. The future is wild in ways I never imagined. Star Trek style holodecks in what, 15 years?

So, creepy thought: should we be recording audio of our parents, so we can still "hear from them" once in a while after they die? People are going to want to reconstruct their lost loved ones with AI. This project seems to imply you only need an hour or so of audio.

Re: 15.ai

#34

Post by jeroenhd »

eljimmy wrote: post #13

This is unrelated but what's with the fascination with HN users and My Little Pony? I've noticed this on a lot of posts in the past few months.

Aside from the causal brony references, this project originally featured a lot of my little pony voices because it needed meticulously annotated transcriptions of the input audio to be trained well.

The extremely dedicated brony subculture voluntarily put in a lot of work to get a corpus for the AI to learn from.

There's also another factor at play: this AI works best with highly pitched voices, which my little pony is just full of. Not only did MLP provide such a generous source of training data, its results were also much more impressive than the dry dictation many other corpi would've resulted in, adding to its fame.

I personally haven't seen any significant rise in MLP references, though that could be because I don't know the show so I don't catch references to it. It's also very possible that you've caught the Baader-Meinhof phenomenon.

Re: 15.ai

#35

Post by BeFlatXIII »

canjobear wrote: post #22
eljimmy wrote: post #13

This is unrelated but what's with the fascination with HN users and My Little Pony? I've noticed this on a lot of posts in the past few months.

A lot of people in tech circles have a sexual fixation on the show and its characters.

It's a good thing they're warehoused in cities and apartments, then.

Re: 15.ai

#38

Post by jeroenhd »

convery wrote: post #23

Interesting how it seems like there's little correlation between source sample-size and quality. e.g. the Portal Sentry turret at 1.5min input vs the 100+ minutes of the narrator from Stanly Parable which sounded like auto-tune had a stroke.

The AI seems to work best on high-pitched, female voices. The model seems to have improved in this regard since I last tried this website, but it's still very significantly biased towards female voices it seems.

Re: 15.ai

#40

Post by deeplearner1 »

If you want more information about 15.ai, I highly suggest reading their Wikipedia article! https://en.wikipedia.org/wiki/15.ai

The whole history behind the project is fascinating: 4chan had a huge role in its development, and the project's work was stolen by an NFT company that a famous voice actor endorsed not too long ago.

Post reply on HN