MacroPad/voice-to-notes
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases 10 Wiki Activity
52 Commits 3 Branches 17 Tags
9cec3c38587662f34e71d7e3a4afcb39208b1e2a
Commit Graph

8 Commits

Author SHA1 Message Date
Claude
58faa83cb3 Cross-platform distribution, UI improvements, and performance optimizations 
- PyInstaller frozen sidecar: spec file, build script, and ffmpeg path resolver
  for self-contained distribution without Python prerequisites
- Dual-mode sidecar launcher: frozen binary (production) with dev mode fallback
- Parallel transcription + diarization pipeline (~30-40% faster)
- GPU auto-detection for diarization (CUDA when available)
- Async run_pipeline command for real-time progress event delivery
- Web Audio API backend for instant playback and seeking
- OpenAI-compatible provider replacing LiteLLM client-side routing
- Cross-platform RAM detection (Linux/macOS/Windows)
- Settings: speaker count hint, token reveal toggles, dark dropdown styling
- Loading splash screen, flexbox layout fix for viewport overflow
- Gitea Actions CI/CD pipeline (Linux, Windows, macOS ARM)
- Updated README and CLAUDE.md documentation

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
 2026-03-20 21:33:43 -07:00
Claude
c23b9a90dd Merge perf/diarize-threading: diarization progress via background thread 2026-03-20 13:52:59 -07:00
Claude
03af5a189c Run pyannote diarization in background thread with progress reporting 
Move the blocking pipeline() call to a daemon thread and emit estimated
progress messages every 2 seconds from the main thread. The progress
estimate uses audio duration to calibrate the expected total time.
Also pass audio_duration_sec from PipelineService to DiarizeService.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
 2026-03-20 13:50:57 -07:00
Josh Knapp
585411f402 Fix speaker diarization: WAV conversion, pyannote 4.0 compat, telemetry bug 
- Convert non-WAV audio to 16kHz mono WAV before diarization (pyannote
  v4.0.4 AudioDecoder returns None duration for FLAC, causing crash)
- Handle pyannote 4.0 DiarizeOutput return type (unwrap .speaker_diarization)
- Disable pyannote telemetry (np.isfinite(None) bug with max_speakers)
- Use huggingface_hub.login() to persist token for all sub-downloads
- Pre-download sub-models (segmentation-3.0, speaker-diarization-community-1)
- Add third required model license link in settings UI
- Improve SpeakerManager hints based on settings state
- Add word-wrap to transcript text

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
 2026-02-26 19:46:07 -08:00
Josh Knapp
baf820286f Add HuggingFace token setting for speaker detection 
- Add "Speakers" tab in Settings with HF token input field
- Include step-by-step instructions for obtaining the token
- Pass hf_token from settings through Rust → Python pipeline → diarize
- Token can also be set via HF_TOKEN environment variable as fallback
- Move skip_diarization checkbox to Speakers tab

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
 2026-02-26 18:08:51 -08:00
Josh Knapp
87b3ad94f9 Improve import UX: progress overlay, pyannote fix, debug logging 
- Enhanced ProgressOverlay with spinner, better styling, and z-index 9999
- Import button shows "Processing..." with pulse animation while transcribing
- Fix pyannote API: use token= instead of deprecated use_auth_token=
- Read HF_TOKEN from environment for pyannote model download
- Add console logging for click-to-seek debugging
- Add color-scheme: dark for native form controls

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
 2026-02-26 17:43:49 -08:00
Josh Knapp
44480906a4 Phase 3: Speaker diarization and full transcription pipeline 
- Implement DiarizeService with pyannote.audio speaker detection
- Build PipelineService combining transcribe → diarize → merge with
  overlap-based speaker assignment per segment
- Add pipeline.start and diarize.start IPC handlers
- Add run_pipeline Tauri command for full pipeline execution
- Wire frontend to use pipeline: speakers auto-created with colors,
  segments assigned to detected speakers
- Build SpeakerManager with rename support (double-click or edit button)
- Add speaker color coding throughout transcript display
- Add pyannote.audio dependency
- Tests: 24 Python (including merge logic), 6 Rust, 0 Svelte errors

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
 2026-02-26 16:09:48 -08:00
Josh Knapp
503cc6c0cf Phase 1 foundation: Tauri shell, Python sidecar, SQLite database 
Tauri v2 + Svelte + TypeScript frontend:
- App shell with workspace layout (waveform, transcript, speakers, AI chat)
- Placeholder components for all major UI areas
- Typed stores (project, transcript, playback, AI)
- TypeScript interfaces matching the database schema
- Tauri bridge service with typed invoke wrappers
- svelte-check passes with 0 errors

Rust backend:
- Tauri v2 app entry point with command registration
- SQLite database layer (rusqlite with bundled SQLite)
  - Full schema: projects, media_files, speakers, segments, words,
    ai_outputs, annotations (with indexes)
  - Model structs with serde serialization
  - CRUD queries for projects, speakers, segments, words
  - Segment text editing preserves original text
  - Schema versioning for future migrations
  - 6 tests passing
- Command stubs for project, transcribe, export, AI, settings, system
- App state management

Python sidecar:
- JSON-line IPC protocol (stdin/stdout)
- Message types: IPCMessage, progress, error, ready
- Handler registry with routing and error handling
- Ping/pong handler for connectivity testing
- Service stubs: transcribe, diarize, pipeline, AI, export
- Provider stubs: local (llama-server), OpenAI, Anthropic, LiteLLM
- Hardware detection stubs
- 14 tests passing, ruff clean

Also adds:
- Testing strategy document (docs/TESTING.md)
- Validation script (scripts/validate.sh)
- Updated .gitignore for Svelte, Rust, Python artifacts

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
 2026-02-26 15:16:06 -08:00