Commit Graph

20 Commits

Author SHA1 Message Date
Cosmo
a885cbe74b feat: VAD-based barge-in during TTS playback 2026-04-14 15:28:14 +00:00
Cosmo
cdf8748e48 feat: VAD-based barge-in during TTS playback 2026-04-14 15:28:12 +00:00
Cosmo
cd921e1540 fix: strip emoji from TTS text in clean_for_speech 2026-04-14 15:26:21 +00:00
3301b3559d Merge branch 'main' of https://git.digital-home.site/daniil/home-voice-assistant 2026-04-14 18:23:41 +03:00
0f4ae3a80c Edit new session 2026-04-14 18:22:01 +03:00
Cosmo
cc9de661cc feat: barge-in support — stop TTS when wake word detected during playback 2026-04-14 15:02:00 +00:00
Cosmo
182e7875ab fix: strip filler phrases from agent response before TTS 2026-04-14 11:49:58 +00:00
a0618c961d Add russian translate 2026-04-14 14:40:52 +03:00
cc8cbefe18 Delete logs 2026-04-14 13:45:38 +03:00
Cosmo
24c8e38be6 fix: replace VOICE_SESSION_KEY with COSMO_SESSION_KEY and LUSYA_SESSION_KEY 2026-04-14 09:45:48 +00:00
09d22177cd Edit voice settings 2026-04-13 23:28:41 +03:00
0494c24c47 Delete conversation from modes 2026-04-13 23:19:18 +03:00
28cccbdac1 Merge pull request 'feat: route voice through OpenClaw agent session (full memory + tools)' (#1) from feature/openclaw-agent-session into main
Reviewed-on: #1
2026-04-13 20:15:02 +00:00
Cosmo
d9d892664a feat: route voice requests through OpenClaw agent session
- Remove local Conversation history (now managed by gateway)
- Use x-openclaw-session-key for persistent agent sessions
- Agent now has full context: SOUL.md, MEMORY.md, tools
- Add VOICE_SESSION_KEY env var (default: agent:main:voice:home)
- Backward compatible: conv parameter kept for compatibility
2026-04-13 20:12:01 +00:00
a836bbb848 Add wakeword cosmo model v1 2026-04-13 18:33:53 +03:00
7239f85506 Edit tts mode 2026-04-13 18:33:19 +03:00
780f6f0084 Switch wake word from Porcupine to openwakeword + training pipeline
- Add training/ pipeline (step_1..step_5) and own-samples flow
- record_wav.py with single-shot and long-record modes, RMS-based silence filter
- remove_silent.py to drop silent samples and renumber
- modes.py: openwakeword inference with reset() and quiet predictions; commented Lusya block for later
- stt.py: drop local faster-whisper fallback, Groq-only
- config.py: remove unused STT_PROVIDER/WHISPER_*
- llm.py: replace __import__("os") hack with proper import
- tts.py: remove debug traceback in play_error_sound
- requirements.txt: add openwakeword/sounddevice/scipy, drop faster-whisper
- deploy/setup.sh: validate ELEVENLABS_API_KEY and WAKE_WORD_COSMO presence
- README.md, CLAUDE.md, project_roadmap memory updated to reflect new architecture
2026-04-13 15:40:44 +03:00
0a89bf5105 Edit code for success run 2026-04-12 21:58:40 +03:00
d.klimov
128cc70ab9 Add voice messages 2026-04-12 15:59:25 +03:00
7ca8268b78 Initial commit: Cosmo Voice Satellite
Two-agent voice assistant (Cosmo + Люся) via OpenClaw Gateway.
Streaming STT (Groq) + LLM + TTS (ElevenLabs) pipeline with
keep-alive sessions, barge-in, and daily conversation sessions.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
2026-04-12 13:34:08 +03:00