Local text to speech with Breeze is truly incredible

A developer has been experimenting with local text-to-speech (TTS) using Breeze, combined with speech-to-text (STT), and found the results to be impressive. Utilizing Opus 5.5, the first sound can be heard after 500ms without processing, or 1-1.5 seconds with low-mode thinking. The developer is considering open-sourcing the project next week and is looking for contributors.

This report details a specific implementation of Breeze TTS with Opus 5.5, providing concrete latency figures unlike general announcements.