built a playground to mess with ai voice infra this week, goal is to see how natural and quick i can get it.

memebuilding • prev manifold.xyz, twilio.com

Check out whisperx

I’ve been running it locally for speech to text and then just using the in browser SpeechSynthesis API

Lots of room to tune but it’s cool to have everything run local

Rated 💯 by multiple networks. Fluent in /unglish. Unbothered by /whatday of the week it is.

Allegedly a designer.

If you’re in Chrome or Safari, there is also the speech recognition api.

feels like everyday I discover a new browser api