Files
claudetools/projects/radio-show/audio-processor/training-data/transcripts/2015-s7e30/diarization.json
Mike Swanson 79abef9dc9 radio: diarization pipeline fixes, benchmark setup, test episode set
- 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>
2026-04-27 13:20:40 -07:00

177 lines
3.2 KiB
JSON

{
"num_speakers": 2,
"speaker_map": {
"CALLER": "CALLER",
"HOST": "HOST"
},
"turns": [
{
"speaker": "CALLER",
"start": 0.0,
"end": 45.0,
"confidence": 0.66
},
{
"speaker": "HOST",
"start": 40.0,
"end": 655.0,
"confidence": 0.92
},
{
"speaker": "CALLER",
"start": 650.0,
"end": 690.0,
"confidence": 0.64
},
{
"speaker": "HOST",
"start": 685.0,
"end": 1350.0,
"confidence": 0.99
},
{
"speaker": "CALLER",
"start": 1345.0,
"end": 1380.0,
"confidence": 0.54
},
{
"speaker": "HOST",
"start": 1375.0,
"end": 1395.0,
"confidence": 0.99
},
{
"speaker": "CALLER",
"start": 1390.0,
"end": 1440.0,
"confidence": 0.84
},
{
"speaker": "HOST",
"start": 1435.0,
"end": 1485.0,
"confidence": 0.87
},
{
"speaker": "CALLER",
"start": 1480.0,
"end": 1495.0,
"confidence": 0.83
},
{
"speaker": "HOST",
"start": 1490.0,
"end": 1515.0,
"confidence": 0.97
},
{
"speaker": "CALLER",
"start": 1510.0,
"end": 1540.0,
"confidence": 0.81
},
{
"speaker": "HOST",
"start": 1535.0,
"end": 1625.0,
"confidence": 0.85
},
{
"speaker": "CALLER",
"start": 1620.0,
"end": 1635.0,
"confidence": 0.81
},
{
"speaker": "HOST",
"start": 1630.0,
"end": 1720.0,
"confidence": 0.96
},
{
"speaker": "CALLER",
"start": 1715.0,
"end": 1725.0,
"confidence": 0.82
},
{
"speaker": "HOST",
"start": 1720.0,
"end": 1860.0,
"confidence": 0.98
},
{
"speaker": "CALLER",
"start": 1855.0,
"end": 1870.0,
"confidence": 0.78
},
{
"speaker": "HOST",
"start": 1865.0,
"end": 2015.0,
"confidence": 0.86
},
{
"speaker": "CALLER",
"start": 2010.0,
"end": 2035.0,
"confidence": 0.82
},
{
"speaker": "HOST",
"start": 2030.0,
"end": 2055.0,
"confidence": 0.97
},
{
"speaker": "CALLER",
"start": 2050.0,
"end": 2070.0,
"confidence": 0.84
},
{
"speaker": "HOST",
"start": 2065.0,
"end": 2075.0,
"confidence": 0.98
},
{
"speaker": "CALLER",
"start": 2070.0,
"end": 2085.0,
"confidence": 0.84
},
{
"speaker": "HOST",
"start": 2080.0,
"end": 2105.0,
"confidence": 0.85
},
{
"speaker": "CALLER",
"start": 2100.0,
"end": 2110.0,
"confidence": 0.77
},
{
"speaker": "HOST",
"start": 2105.0,
"end": 2345.0,
"confidence": 0.96
},
{
"speaker": "CALLER",
"start": 2340.0,
"end": 2390.0,
"confidence": 0.68
},
{
"speaker": "HOST",
"start": 2385.0,
"end": 2720.0,
"confidence": 0.92
}
]
}