Speaker diarization via pyannote #1
Labels
No milestone
No project
No assignees
1 participant
Notifications
Due date
No due date set.
Dependencies
No dependencies set.
Reference
Mark/whisper-transcribe#1
Loading…
Add table
Add a link
Reference in a new issue
No description provided.
Delete branch "%!s()"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
WhisperX's headline feature. Two-model pipeline: speaker-diarization-3.1 + segmentation-3.0.
State: EULAs already accepted for both models in browser session. Blocked only on first-run HF token.
What needs building:
--diarizeflag on the CLI{speaker: 'SPEAKER_00', start: 12.3, end: 14.1, text: '...'})HF_TOKENmissing or pyannote weights not cachedReference: the Lex Fridman 3.4-hour Tao episode (
lex_long.wav) is a good two-speaker test — host + Tao. Should produceSPEAKER_00/SPEAKER_01segments.Triaged and unblocked: HF token is now in place at ~/.huggingface/token (rotated 2026-09-06). pyannote weights should download without the auth bootstrap failure. Ready for the kilo-gsd lane.