-
v1.6 Voice Pipeline + Minimal Message Bridge
Some checks failedDocker / build-and-push (push) Has been cancelledreleased this
2026-04-04 03:52:33 +00:00 | -536 commits to PAP-878-create-a-mine-tab-in-inbox since this releaseDelivered: Transport-agnostic voice pipeline (Whisper STT + Piper TTS) with web chat voice UI, minimal Telegram bridge, and sentence-buffered streaming.
Key accomplishments:
- VoicePipelineService: transcribe, synthesize, formatForVoice with ffmpeg transcoding
- Web chat: VAD silence detection, waveform, voice mode toggle, inline audio playback
- Telegram bridge: grammY long polling, text+voice relay, BotFather onboarding
- Sentence-buffered TTS streaming and multi-language synthesis
- Onboarding hardware probe for STT/TTS capability detection
4 phases, 12 plans, 23 requirements satisfied.
See .planning/MILESTONES.md for full details.Downloads
-
Source code (ZIP)
1 download
-
Source code (TAR.GZ)
1 download