3 Commits

Author	SHA1	Message	Date
Ethanfel	8ab5bdba77	fix: use mean+std MFCC vectors (40-dim) for better discrimination ... Mean-only vectors were too similar across different audio segments, causing everything to match even at threshold 0.99. Adding std captures temporal dynamics and makes the similarity scores much more spread out. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>	2026-04-17 09:27:11 +02:00
Ethanfel	9cf9e3233f	feat: add scan_video with average and nearest modes ... Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>	2026-04-17 08:50:47 +02:00
Ethanfel	e17d8f67aa	feat: add audio_scan module with build_profile ... Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>	2026-04-17 08:48:18 +02:00