Files
claudetools/projects/radio-show/audio-processor/test-data/transcripts/2011-03-12-hr1/diarization.json
Mike Swanson a4f527f31e radio: per-year test set (one episode per year, 2010-2018)
Added 2010, 2015, 2018 test episodes to round out the test set to one
per available year:
- 2010-05-08-hr1 (May 2010, earliest available; pre-Tara era)
- 2015-s7e19 (Jan 2015, avoids training's s7e30)
- 2018-s10e18 (only 3 non-training 2018 episodes exist)

Archive has no 2019 directory — Rob's "2018/2019 appearances" are
constrained to the 5 available 2018 episodes only.

Per-year diarization summary (Tara presence, post-rename):
  2010-05-08    30s   1.2%   likely false positive (pre-Tara)
  2011-03-12   140s   5.6%   likely false positive (call-in only)
  2012-03-10    30s   1.1%   likely false positive (call-in only)
  2012-06-09   340s  12.8%   suspicious — Mike to confirm
  2014-s6e19   680s  23.3%   confirmed
  2015-s7e19   280s   9.9%   plausible — Mike to confirm
  2016-s8e43  1890s  35.5%   confirmed
  2017-s9e30   610s  11.4%   plausible
  2018-s10e18  880s  17.1%   COULD BE ROB — Mike flagged Rob for
                              2018/2019 appearances; cosine threshold may
                              be hitting on Rob being acoustically similar
                              to Tara

Total Tara across 9 episodes: 1h 21m / 8h 52m audio (15.3%).

Q&A counts (still suspect — every voice that isn't Mike-or-Tara is
labeled CALLER, so Randall/Rob/producers inflate the bucket):
  2010=4, 2011=1, 2012a=2, 2012b=0, 2014=0, 2015=1, 2016=2, 2017=4, 2018=3
  Total: 17 pairs across 9 episodes

4090 perf on the expanded set:
- Diarization: 31928s in 121.5s = 262.7x realtime (vs 209.7x on 5070 Ti, +25.3%)
- Transcription (3 new episodes only): 10554s in 112.4s = 93.9x

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-04-27 15:20:09 -07:00

160 lines
2.9 KiB
JSON

{
"num_speakers": 3,
"speaker_map": {
"CALLER": "CALLER",
"HOST": "HOST",
"CO-HOST": "CO-HOST"
},
"turns": [
{
"speaker": "HOST",
"start": 0.0,
"end": 20.0,
"confidence": 0.89
},
{
"speaker": "CO-HOST",
"start": 15.0,
"end": 25.0,
"confidence": 0.87
},
{
"speaker": "HOST",
"start": 20.0,
"end": 40.0,
"confidence": 0.88
},
{
"speaker": "CALLER",
"start": 35.0,
"end": 45.0,
"confidence": 0.83
},
{
"speaker": "HOST",
"start": 40.0,
"end": 515.0,
"confidence": 0.9
},
{
"speaker": "CO-HOST",
"start": 510.0,
"end": 520.0,
"confidence": 0.9
},
{
"speaker": "HOST",
"start": 515.0,
"end": 925.0,
"confidence": 0.97
},
{
"speaker": "CALLER",
"start": 920.0,
"end": 935.0,
"confidence": 0.82
},
{
"speaker": "HOST",
"start": 930.0,
"end": 1000.0,
"confidence": 0.89
},
{
"speaker": "CALLER",
"start": 995.0,
"end": 1005.0,
"confidence": 0.81
},
{
"speaker": "HOST",
"start": 1000.0,
"end": 1205.0,
"confidence": 0.95
},
{
"speaker": "CALLER",
"start": 1200.0,
"end": 1215.0,
"confidence": 0.81
},
{
"speaker": "HOST",
"start": 1210.0,
"end": 1280.0,
"confidence": 0.97
},
{
"speaker": "CALLER",
"start": 1275.0,
"end": 1285.0,
"confidence": 0.81
},
{
"speaker": "HOST",
"start": 1280.0,
"end": 1485.0,
"confidence": 0.91
},
{
"speaker": "CO-HOST",
"start": 1480.0,
"end": 1500.0,
"confidence": 0.97
},
{
"speaker": "HOST",
"start": 1495.0,
"end": 1775.0,
"confidence": 0.98
},
{
"speaker": "CO-HOST",
"start": 1770.0,
"end": 1785.0,
"confidence": 0.91
},
{
"speaker": "HOST",
"start": 1780.0,
"end": 1835.0,
"confidence": 0.98
},
{
"speaker": "CALLER",
"start": 1830.0,
"end": 1840.0,
"confidence": 0.84
},
{
"speaker": "HOST",
"start": 1835.0,
"end": 2160.0,
"confidence": 0.95
},
{
"speaker": "CO-HOST",
"start": 2155.0,
"end": 2175.0,
"confidence": 0.87
},
{
"speaker": "HOST",
"start": 2170.0,
"end": 2295.0,
"confidence": 0.96
},
{
"speaker": "CO-HOST",
"start": 2290.0,
"end": 2315.0,
"confidence": 0.93
},
{
"speaker": "HOST",
"start": 2310.0,
"end": 2505.0,
"confidence": 0.98
}
]
}