Speakrs
Model Serving & MLOpsFast Rust speaker diarization library achieving pyannote‑level accuracy with CPU, GPU, or CoreML backends
Features
- Full pyannote community‑1 style diarization pipeline implemented in Rust
- Supports multiple execution modes: ONNX Runtime CPU, CUDA/GPU, AMD MIGraphX, and native macOS CoreML
- Offers both high‑precision (1 s window) and higher‑throughput (2 s window) modes
- Provides queued background processing for streaming audio workloads
Recent releases
View all 2 releases →Weekly OSS security release digest.
The CVE patches and breaking changes that affected production tools this week. One email, every Sunday.
No spam, unsubscribe anytime.
Install & Platforms
Install via
cargo
Platforms
macos
linux
windows
arm64
Similar tools
Alternative to
pyannote
pyannote-rs