Overview - Naxodev/mar

naxodev/mar

7 days30 days

Latest CI Pipeline Executions

Fuzzy

Succeeded
main
ddbd47eb refactor: client/voice (audio playback TTS) (#16) ## PR Type What kind of change does this PR introduce?  ``` [ ] Bugfix [ ] Feature [ ] Code style update (formatting, local variables) [x] Refactoring (no functional changes, no api changes) [ ] Build related changes [ ] CI related changes [ ] Documentation content changes [ ] Other... Please describe: ``` ## What is the current behavior? The voice functionality is currently tightly coupled in the React layer with: - Audio capture logic embedded in `use-maelstrom-voice.ts` (using inline AudioWorklet code) - Audio playback handling mixed with state management - VAD logic partially in React hooks - Complex orchestration spread across multiple files - Limited test coverage for audio operations ## What is the new behavior? This PR refactors the voice architecture by extracting concerns into dedicated packages: 1. **`audio-capture`**: New standalone package with `AudioCapture` class handling microphone access, AudioWorklet management, and audio streaming 2. **`audio-playback`**: New standalone package with `AudioPlayback` class for TTS audio queue management and scheduling 3. **`voice-session`**: New package containing `VoiceSession` class that orchestrates STT/TTS flow, VAD integration, and state management 4. **Simplified React hooks**: `use-maelstrom-voice.ts` reduced from ~800 lines to ~200 lines by delegating to `VoiceSession` 5. **Improved VAD**: Better Silero VAD integration with configurable timeout options 6. **Enhanced testing**: Added comprehensive test suites for audio-capture (328 lines), audio-playback (194 lines), and voice-session (1238 lines) 7. **Better concurrency**: Fixed race conditions in audio playback and capture 8. **Code organization**: Moved all audio processing logic from React layer to client library API remains unchanged for consumers of the React hook. ## Does this PR introduce a breaking change? ``` [ ] Yes [x] No ```  ## Other information - Removed ~700 lines from `use-maelstrom-voice.ts` and ~777 lines from `voice-connector.ts` - Total net addition of ~1,300 lines (mostly new tests and documentation) - All existing voice functionality preserved - Improves maintainability by separating audio concerns from React UI logic - Enables reuse of voice logic in non-React contexts
16 hours ago by Javier Cab...
Succeeded
refactoring/client/voice
b90ab012 adding max user speech as option in vad fix
19 hours ago by Javier Cab...
Failed
refactoring/client/voice
Fix ready→
55b9fd68 adding max user speech as option in vad
19 hours ago by Javier Cab...
Failed
refactoring/client/voice
Fix ready→
a66595e8 long hard limit
19 hours ago by Javier Cab...
Failed
refactoring/client/voice
Fix ready→
83055c82 fix vad sould be able to decide if interim is final after timeout fix
19 hours ago by Javier Cab...
Failed
refactoring/client/voice
Fix ready→
3e2f8fd4 lab results
20 hours ago by Javier Cab...
Failed
refactoring/client/voice
Fix ready→
49aefd70 nx suggestion
3 days ago by Jacarte
Failed
refactoring/client/voice
b125d31b fix
3 days ago by Jacarte
Failed
refactoring/client/voice
Fix applied→
01a32b10 linting
3 days ago by Jacarte
Failed
refactoring/client/voice
d3a7ea28 no more Set<Function>
3 days ago by Jacarte
Failed
refactoring/client/voice
Fix ready→
18e4c209 private methods solution
3 days ago by Jacarte
Failed
refactoring/client/voice
Fix ready→
a64b11e4 concurrency fixes
3 days ago by Jacarte
Failed
refactoring/client/voice
Fix ready→
bad8b003 fine tunning
3 days ago by Jacarte
Succeeded
main
467d7eb7 fix: zero executor not found
5 days ago by NachoVazquez
Failed
refactoring/client/voice
Fix ready→
911e42f7 merging
5 days ago by Jacarte
Succeeded
main
d9b89d27 fix: forward client VAD events to voice-engine and use them to finalize transcripts (#15) Description: --- ## Summary - Forward client-side VAD (speech_start/speech_end) events into the voice-engine via a new handleVADEvent method, so the engine is aware of speech boundaries detected in the browser - Use VAD speech_end to mark the next interim transcript as final, with a 200ms debounce fallback if no new interim arrives — prevents Deepgram from hanging on open-ended streams - Fix STT-provider VAD events (from Deepgram) that were logged but never propagated to listeners/callbacks ## Changes engine/src/lib/voice/types.ts - Added handleVADEvent(event: VADEvent) to the VoiceEngine interface engine/src/lib/voice/voice-engine.ts - Implemented handleVADEvent: propagates to listeners/callbacks, and on speech_end sets a flag so the next interim is promoted to final (or debounces at 200ms to finalize the last known transcript) - Fixed handleSTTEvent vad case: now propagates to vadListeners and callbacks.onVADEvent instead of just logging - Added hasEmittedFinal guard to prevent duplicate final transcripts between the VAD path and stopListening() mar/maelstrom/server/api/src/lib/maelstrom-ws.ts - Added voiceEngine.handleVADEvent() calls in both vad.speech_start and vad.speech_end handlers, alongside the existing inline barge-in + start/stop listening logic (preserved for recording and Deepgram stream management)
5 days ago by Jacarte
Succeeded
naxo/mae-52-extract-vessel-management-subdomain
e43f4639 feat: integrate with the client v1 and react
6 days ago by NachoVazquez
Succeeded
naxo/mae-51-extract-connection-management-subdomain
b0f22849 fix: type issues
6 days ago by NachoVazquez
Failed
naxo/mae-51-extract-connection-management-subdomain
Fix ready→
a180cdc7 refactor: more state renamed to status
6 days ago by NachoVazquez
Succeeded
naxo/mae-51-extract-connection-management-subdomain
0dc889b8 docs: add jsdocs to connection manager
7 days ago by NachoVazquez