Results for polyvoice

Speaker diarization for Python — who spoke when. Rust + ONNX, no Python runtime overhead. K-means/AHC clustering, overlap detection.

Speaker diarization for Rust — who spoke when. ONNX-powered: Silero VAD, WeSpeaker embeddings, Pyannote segmentation, K-means/AHC clustering, overlap detection.

Core inference engine for gigastt — GigaAM v3 ONNX Runtime, model management, quantization

Local STT server powered by Zipformer-vi RNN-T — on-device Vietnamese speech recognition via ONNX Runtime