polyvoice 0.6.9

Speaker diarization for Rust — who spoke when. ONNX-powered: Silero VAD, WeSpeaker embeddings, Pyannote segmentation, K-means/AHC clustering, overlap detection.
Documentation 
1
2
3
4
5
6
7
8
9
10
11
12
13
14

# TODO — src/segmentation

## Current

## Next

- [ ] Add property test: decoder never panics on arbitrary logits.
- [ ] Benchmark Aggregator Hungarian vs greedy fallback.

## Known Gaps

## Deferred

- [ ] Explore streaming segmentation variant.