avencera/speakrs

Speaker diarization in Rust with pyannote-level accuracy. 1790–5970x realtime on NVIDIA GPUs, 430–910x on Apple Silicon.

语言：Rust · GitHub 星标：182 · Fork：28

话题：audio · coreml · cuda · diarization · ml · nvidia · onnx-runtime · pyannote

在 GitHub 上查看原仓库