Files
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

393 lines
7.2 KiB
JSON

{
"num_speakers": 2,
"speaker_map": {
"CALLER": "CALLER",
"HOST": "HOST"
},
"turns": [
{
"speaker": "CALLER",
"start": 0.0,
"end": 40.0,
"confidence": 0.68
},
{
"speaker": "HOST",
"start": 35.0,
"end": 380.0,
"confidence": 0.98
},
{
"speaker": "CALLER",
"start": 375.0,
"end": 385.0,
"confidence": 0.84
},
{
"speaker": "HOST",
"start": 380.0,
"end": 685.0,
"confidence": 0.96
},
{
"speaker": "CALLER",
"start": 680.0,
"end": 700.0,
"confidence": 0.61
},
{
"speaker": "HOST",
"start": 695.0,
"end": 705.0,
"confidence": 0.87
},
{
"speaker": "CALLER",
"start": 700.0,
"end": 710.0,
"confidence": 0.84
},
{
"speaker": "HOST",
"start": 705.0,
"end": 1235.0,
"confidence": 0.96
},
{
"speaker": "CALLER",
"start": 1230.0,
"end": 1240.0,
"confidence": 0.84
},
{
"speaker": "HOST",
"start": 1235.0,
"end": 1365.0,
"confidence": 0.91
},
{
"speaker": "CALLER",
"start": 1360.0,
"end": 1410.0,
"confidence": 0.79
},
{
"speaker": "HOST",
"start": 1405.0,
"end": 1430.0,
"confidence": 0.95
},
{
"speaker": "CALLER",
"start": 1425.0,
"end": 1435.0,
"confidence": 0.78
},
{
"speaker": "HOST",
"start": 1430.0,
"end": 1440.0,
"confidence": 0.86
},
{
"speaker": "CALLER",
"start": 1435.0,
"end": 1455.0,
"confidence": 0.82
},
{
"speaker": "HOST",
"start": 1450.0,
"end": 1475.0,
"confidence": 0.96
},
{
"speaker": "CALLER",
"start": 1470.0,
"end": 1480.0,
"confidence": 0.77
},
{
"speaker": "HOST",
"start": 1475.0,
"end": 2155.0,
"confidence": 0.97
},
{
"speaker": "CALLER",
"start": 2150.0,
"end": 2185.0,
"confidence": 0.82
},
{
"speaker": "HOST",
"start": 2180.0,
"end": 2195.0,
"confidence": 0.89
},
{
"speaker": "CALLER",
"start": 2190.0,
"end": 2205.0,
"confidence": 0.83
},
{
"speaker": "HOST",
"start": 2200.0,
"end": 2285.0,
"confidence": 0.85
},
{
"speaker": "CALLER",
"start": 2280.0,
"end": 2295.0,
"confidence": 0.84
},
{
"speaker": "HOST",
"start": 2290.0,
"end": 2505.0,
"confidence": 0.86
},
{
"speaker": "CALLER",
"start": 2500.0,
"end": 2510.0,
"confidence": 0.79
},
{
"speaker": "HOST",
"start": 2505.0,
"end": 2535.0,
"confidence": 0.98
},
{
"speaker": "CALLER",
"start": 2530.0,
"end": 2565.0,
"confidence": 0.45
},
{
"speaker": "HOST",
"start": 2560.0,
"end": 2665.0,
"confidence": 0.98
},
{
"speaker": "CALLER",
"start": 2660.0,
"end": 2670.0,
"confidence": 0.84
},
{
"speaker": "HOST",
"start": 2665.0,
"end": 3040.0,
"confidence": 0.97
},
{
"speaker": "CALLER",
"start": 3035.0,
"end": 3045.0,
"confidence": 0.85
},
{
"speaker": "HOST",
"start": 3040.0,
"end": 3360.0,
"confidence": 0.9
},
{
"speaker": "CALLER",
"start": 3355.0,
"end": 3370.0,
"confidence": 0.84
},
{
"speaker": "HOST",
"start": 3365.0,
"end": 3385.0,
"confidence": 0.86
},
{
"speaker": "CALLER",
"start": 3380.0,
"end": 3395.0,
"confidence": 0.85
},
{
"speaker": "HOST",
"start": 3390.0,
"end": 3475.0,
"confidence": 0.92
},
{
"speaker": "CALLER",
"start": 3470.0,
"end": 3480.0,
"confidence": 0.83
},
{
"speaker": "HOST",
"start": 3475.0,
"end": 3520.0,
"confidence": 0.97
},
{
"speaker": "CALLER",
"start": 3515.0,
"end": 3550.0,
"confidence": 0.59
},
{
"speaker": "HOST",
"start": 3545.0,
"end": 3555.0,
"confidence": 0.98
},
{
"speaker": "CALLER",
"start": 3550.0,
"end": 3660.0,
"confidence": 0.56
},
{
"speaker": "HOST",
"start": 3655.0,
"end": 3695.0,
"confidence": 0.97
},
{
"speaker": "CALLER",
"start": 3690.0,
"end": 3705.0,
"confidence": 0.49
},
{
"speaker": "HOST",
"start": 3700.0,
"end": 3730.0,
"confidence": 0.87
},
{
"speaker": "CALLER",
"start": 3725.0,
"end": 3750.0,
"confidence": 0.6
},
{
"speaker": "HOST",
"start": 3745.0,
"end": 3800.0,
"confidence": 0.85
},
{
"speaker": "CALLER",
"start": 3795.0,
"end": 3810.0,
"confidence": 0.57
},
{
"speaker": "HOST",
"start": 3805.0,
"end": 3900.0,
"confidence": 0.96
},
{
"speaker": "CALLER",
"start": 3895.0,
"end": 3905.0,
"confidence": 0.46
},
{
"speaker": "HOST",
"start": 3900.0,
"end": 4080.0,
"confidence": 0.98
},
{
"speaker": "CALLER",
"start": 4075.0,
"end": 4085.0,
"confidence": 0.82
},
{
"speaker": "HOST",
"start": 4080.0,
"end": 4210.0,
"confidence": 0.87
},
{
"speaker": "CALLER",
"start": 4205.0,
"end": 4250.0,
"confidence": 0.65
},
{
"speaker": "HOST",
"start": 4245.0,
"end": 4595.0,
"confidence": 0.97
},
{
"speaker": "CALLER",
"start": 4590.0,
"end": 4600.0,
"confidence": 0.84
},
{
"speaker": "HOST",
"start": 4595.0,
"end": 4765.0,
"confidence": 0.88
},
{
"speaker": "CALLER",
"start": 4760.0,
"end": 4830.0,
"confidence": 0.49
},
{
"speaker": "HOST",
"start": 4825.0,
"end": 4885.0,
"confidence": 0.95
},
{
"speaker": "CALLER",
"start": 4880.0,
"end": 4910.0,
"confidence": 0.56
},
{
"speaker": "HOST",
"start": 4905.0,
"end": 4940.0,
"confidence": 0.97
},
{
"speaker": "CALLER",
"start": 4935.0,
"end": 4975.0,
"confidence": 0.52
},
{
"speaker": "HOST",
"start": 4970.0,
"end": 5125.0,
"confidence": 0.97
},
{
"speaker": "CALLER",
"start": 5120.0,
"end": 5130.0,
"confidence": 0.81
},
{
"speaker": "HOST",
"start": 5125.0,
"end": 5420.0,
"confidence": 0.91
}
]
}