- Fix voice_profiler threshold bug (HOST label overwrote Unknown unconditionally) - Audio preload optimization: single ffmpeg per episode, 149.5x realtime on 5070 Ti - WavLM threshold raised to 0.85 (Mike 0.90-0.99, callers 0.46-0.83) - Promo/bumper filter: weighted signature scoring, 42->27 clean Q&A pairs - Text-only Q&A fallback for episodes with no CALLER diarization labels - TRANSFORMERS_OFFLINE=1 to skip HuggingFace freshness checks - Add diarize_2018.py for targeted re-run + FTS5 rebuild - Add benchmark.py + BENCH_SETUP.md for GURU-BEAST-ROG (RTX 4090) comparison - Commit 9-episode training diarization.json outputs - Session log: 2026-04-27-diarization-pipeline.md Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
111 lines
2.0 KiB
JSON
111 lines
2.0 KiB
JSON
{
|
|
"num_speakers": 2,
|
|
"speaker_map": {
|
|
"HOST": "HOST",
|
|
"CALLER": "CALLER"
|
|
},
|
|
"turns": [
|
|
{
|
|
"speaker": "HOST",
|
|
"start": 0.0,
|
|
"end": 20.0,
|
|
"confidence": 0.86
|
|
},
|
|
{
|
|
"speaker": "CALLER",
|
|
"start": 15.0,
|
|
"end": 35.0,
|
|
"confidence": 0.84
|
|
},
|
|
{
|
|
"speaker": "HOST",
|
|
"start": 30.0,
|
|
"end": 475.0,
|
|
"confidence": 0.94
|
|
},
|
|
{
|
|
"speaker": "CALLER",
|
|
"start": 470.0,
|
|
"end": 495.0,
|
|
"confidence": 0.79
|
|
},
|
|
{
|
|
"speaker": "HOST",
|
|
"start": 490.0,
|
|
"end": 635.0,
|
|
"confidence": 0.94
|
|
},
|
|
{
|
|
"speaker": "CALLER",
|
|
"start": 630.0,
|
|
"end": 640.0,
|
|
"confidence": 0.82
|
|
},
|
|
{
|
|
"speaker": "HOST",
|
|
"start": 635.0,
|
|
"end": 720.0,
|
|
"confidence": 0.95
|
|
},
|
|
{
|
|
"speaker": "CALLER",
|
|
"start": 715.0,
|
|
"end": 725.0,
|
|
"confidence": 0.62
|
|
},
|
|
{
|
|
"speaker": "HOST",
|
|
"start": 720.0,
|
|
"end": 1310.0,
|
|
"confidence": 0.87
|
|
},
|
|
{
|
|
"speaker": "CALLER",
|
|
"start": 1305.0,
|
|
"end": 1320.0,
|
|
"confidence": 0.7
|
|
},
|
|
{
|
|
"speaker": "HOST",
|
|
"start": 1315.0,
|
|
"end": 1565.0,
|
|
"confidence": 0.86
|
|
},
|
|
{
|
|
"speaker": "CALLER",
|
|
"start": 1560.0,
|
|
"end": 1570.0,
|
|
"confidence": 0.68
|
|
},
|
|
{
|
|
"speaker": "HOST",
|
|
"start": 1565.0,
|
|
"end": 2055.0,
|
|
"confidence": 0.97
|
|
},
|
|
{
|
|
"speaker": "CALLER",
|
|
"start": 2050.0,
|
|
"end": 2060.0,
|
|
"confidence": 0.85
|
|
},
|
|
{
|
|
"speaker": "HOST",
|
|
"start": 2055.0,
|
|
"end": 2080.0,
|
|
"confidence": 0.92
|
|
},
|
|
{
|
|
"speaker": "CALLER",
|
|
"start": 2075.0,
|
|
"end": 2090.0,
|
|
"confidence": 0.76
|
|
},
|
|
{
|
|
"speaker": "HOST",
|
|
"start": 2085.0,
|
|
"end": 2560.0,
|
|
"confidence": 0.85
|
|
}
|
|
]
|
|
} |