Files
hermes-agent/hermes_cli
brooklyn! 20bcc9bd14 fix(tts): stream Edge speak-stream per sentence instead of whole-text fallback
Desktop speak-stream sent type=fallback whenever the provider had no
chunked PCM API. Edge is that case, so the client waited for the full
reply and POSTed it. Cut sentences with the existing sync TTS tool and
stream that PCM. Fallback stays the last resort when synthesis produces
no audio.

Refs #91997
2026-09-24 19:28:05 -05:00
..
…
…
…
…
…
…
…