- Add speech detection via faster-whisper with red waveform coloring for speech regions
- Add format variant export buttons (P/S) next to Export and subprofile buttons when portrait/square enabled
- Add force_ratio parameter to _on_export for deterministic format exports
- Add subcategory show/hide with persistent checkbox menu (no longer closes on toggle)
- Show crop overlay lines during video playback, not just when paused
- Delete marker now also removes files from disk and cleans up annotations
- Clear all markers also deletes files and DB entries
- Add playlist text filter, clip spread tick lines on timeline
- Fix LD_PRELOAD for GLIBCXX in conda launcher
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- Replace librosa with direct ffmpeg subprocess for 10x faster audio loading
- Add ScanResultsPanel with per-model tabs, seek-on-click, delete, and export
- Persist scan results in DB per (filename, profile, model)
- Add model selector dropdown to switch between trained embedding models
- Add "Scan All" button for batch scanning playlist videos
- Support manual negative examples via negative class folder
- Configurable auto-negative margin (default 30s, 0 = disabled)
- Deduplicate nearby training markers (8s min gap)
- Parallel audio loading with ThreadPoolExecutor during training
- Progress callbacks from training for UI status updates
- Cache bypass in scan_video (skip audio loading when embeddings cached)
- Move all caches (models, embeddings, downloads) into project directory
- Add 8cut.sh launcher script with auto venv/conda detection
- Fix 11 bugs across thread safety, signal handling, and state management
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Ethanfel