(summary) improve speaker assignment

Speaker-to-participant assignment relie on WhisperX word timings, but
incorrect word durations in the output can lead to inaccurate overlap
scoring and wrong user attribution. Add a custom heuristic to trim
overly long word durations before computing assignments.
This commit is contained in:
leo
2026-05-12 16:11:46 +02:00
committed by aleb_the_flash
parent 02d16cb55c
commit 96f97ed2d0
4 changed files with 302 additions and 15 deletions
+1
View File
@@ -23,6 +23,7 @@ and this project adheres to
### Changed
- 🧑‍💻(agents) use `uv` for package management
- ✨(summary) improve speaker-to-participant assignment
### Fixed