Speech-to-speech models: the end of the text middleman?

Skip transcription and generation as separate steps and latency drops dramatically.

511

The hard part isn't the AI, it's trust.

Following up: your tips worked. Marking this as done ✅

cfontaine, ifeN, Isabela_O26 and 10 others like this.

Hot take: speech-to-speech models is the most underrated part of XIRO. ✨

Callum_H84, julesF, meera-g and 1 other like this.

This deserves more attention. Speech-to-speech models will matter a lot next year.

172

@hannah-t Can you share your setup? I'd like to try speech-to-speech models too. 🔥

@zeynepA Remember when we said touchscreens were a fad? 😉