82940d96d7
radio: utf-8 transcript writes + sqlite archive importer + session log
...
- src/transcriber.py: open transcript.{json,txt,srt} with encoding="utf-8".
Windows cp1252 default crashed on Whisper output containing U+2044.
- import_to_sqlite.py: new. Walks archive-data/transcripts, builds
archive.db (5 tables + 2 FTS5 virtual tables, sha256-keyed idempotency).
20.5 MB / 208 episodes at smoke-test time, 1.9s rebuild.
- batch_process.py: tracked from prior session — full-archive batch with
resumable transcribe/diarize/intros/qa pipeline.
- .gitignore: archive-data/ and logs/.
Session log: 2026-04-27-archive-batch-and-sqlite-import.md.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com >
2026-04-27 19:38:02 -07:00
e9ac607500
radio show: co-host voice profile, Q&A extraction fixes, archive index
...
- Build Tom (co-host) voice profile (44 embeddings, 0.698 similarity to Mike)
- diarizer.py: add CO-HOST speaker label for cohost-role profiles
- voice_profiler.py: emit "Cohost: <name>" label for cohost role
- qa_extractor.py: overlap resolution at load time (midpoint boundary split),
4s CALLER-preference threshold, turn-based caller-intro lookback (2 HOST turns),
_preceded_by_caller_intro() helper, _PHONE_GREETING pattern,
751-1041 + "we'll get your problem solved" promo signatures
- benchmark.py: use src.transcriber.transcribe with batch_size=16
- add index_test_episodes.py and build_cohost_profile.py scripts
- add .gitignore (exclude episodes, transcripts, *.db, .venv)
- session log: 2026-04-27-qa-extraction-cohost-indexing.md
Result: 2016-s8e43 drops from 12 false-positive Q&A pairs to 2 real caller pairs.
archive.db: 6 episodes, 762 segments, 10 Q&A pairs, FTS5 search verified.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com >
2026-04-27 14:41:04 -07:00
a1e0442d8b
Add radio show audio processor and post-show workflow
...
- Audio processor CLI tool with 6-stage pipeline: transcribe (faster-whisper GPU),
diarize (pyannote), detect segments (multi-signal classifier), remove commercials,
split segments, analyze content (Ollama)
- Post-show workflow doc for episode posts, forum threads, deep-dive blog posts
- Training plan for using 579-episode archive for voice profiles and commercial detection
- Successful test: 45min episode transcribed in 2:37 on RTX 5070 Ti
- Sample transcript output from S7E30 (March 2015)
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com >
2026-03-21 11:51:59 -07:00
