Files
pyannote-speaker-diarizatio…/docs/parity-evidence.md
transcrilive 2b1a3c1312 feat: initial public release v0.1.0 — MLX port of pyannote-speaker-diarization-3.1
Byte-parity with pyannote-PyTorch reference (cosine 0.763718 identical
at 6 decimals on 200 cross-window slot pairs). 2.5x faster than
pyannote-MPS on Apple Silicon native.

Extracted from gitea.tavportal.com/olivier/MLX_CONVERTOR commit 5f9eafa.
2026-05-09 16:05:39 +02:00

437 B

Parity Evidence

Evidence MLX Reference Result
Cosine distance parity 200 cross-window pairs, mean 0.763718 pyannote-PyTorch mean 0.763718 identical at 6 decimals
5h10 bench results 173s wall / 44 speakers / 1.27 GB peak RSS pyannote-MPS 431s / 43 speakers / 1.72 GB Cross-DER 0.076
Source commits 8aa6c6d + 5f9eafa feat/platform-abc in MLX_CONVERTOR extraction source