Hunter Carter

Where is the repo?

Link is broken

Excellent release guys. This is literally the most exciting development in the OSS / AI field this year, for me. Can't wait to see how you grow this ecosystem.

As a Hollywood professional and freelancer the name is a big turn off. It speaks to hustle culture, exactly the thing that most people who have gone freelance left their 9-to-5 to avoid.

You're not going to be able to keep that name, and the misappropriation of it does not inspire confidence in your judgement.

Disingenuous to advertise as open source when it is only the SDK that is OSS, and the platform is closed.

How did you achieve the real time image generation, from a technical/devops pov?

Did you license these voices?

To what degree does the fact that the content was previously unreleased matter? Isn't any "new" content also previously unreleased, and wouldn't this serve the same function, regardless of how long the content been on the shelf?

Feels like a joke to search for mentors and see 90% 20-somethings.

Dystopian.

LOVE THIS. Thank you for open-sourcing it. What's the ETA for the features mentioned on the landing page?

The road to hell is paved with good intentions.

Seems unethical to tell people they're compatible or not based on a single point of data, with a technology known to hallucinate. Your app, as currently conceived, will probably end relationships.

Folsom is the name of a prison. Most people who know the word would think of that first, because it was made popular by Johnny Cash. Strange name choice.

How the hell do you expect to get away with using the Yoda name? Disney owns it.

You should honestly be testing value proposition copy before pulling the trigger on any product stuff. So, it follows, that the copy writing cycle lasts for the entire sprint.

I could have sworn that I heard they killed this on account of the fact that it makes everyone shit their pants?

AI alignment and truth decay (they go together).

What model are you using for the lipsync and what resolution output do you support? I am aware of the wav2lip model, but AFAIK its lip rendering window is limited to something like 128x128. Can this do any higher?