pyannote-rs
| Crates.io | pyannote-rs |
| lib.rs | pyannote-rs |
| version | 0.2.7 |
| source | src |
| created_at | 2024-08-06 15:57:20.209334 |
| updated_at | 2024-08-18 20:10:42.939486 |
| description | Speaker diarization using pyannote in Rust |
| homepage | |
| repository | |
| max_upload_size | |
| id | 1327464 |
| size | 47,310 |
(thewh1teagle)
documentation
README
pyannote-rs
Pyannote audio diarization in Rust
Features
- Compute 1 hour of audio in less than a minute on CPU.
- Faster performance with DirectML on Windows and CoreML on macOS.
- Accurate timestamps with Pyannote segmentation.
- Identify speakers with wespeaker embeddings.
Install
cargo add pyannote-rs
Usage
See Building
Examples
See examples
How it works
pyannote-rs uses 2 models for speaker diarization:
- Segmentation: segmentation-3.0 identifies when speech occurs.
- Speaker Identification: wespeaker-voxceleb-resnet34-LM identifies who is speaking.
Inference is powered by onnxruntime.
- The segmentation model processes up to 10s of audio, using a sliding window approach (iterating in chunks).
- The embedding model processes filter banks (audio features) extracted with knf-rs.
Speaker comparison (e.g., determining if Alice spoke again) is done using cosine similarity.
Credits
Big thanks to pyannote-onnx and kaldi-native-fbank