Silero Vad Python Packages

senko

Very fast, accurate speaker diarization

cohere-transcribe

Production-ready batch inference for Cohere’s Arabic/English ASR model, with optimized Silero VAD, multi-file GPU batching, subtitles, and optional word-level timestamps.

live-translation

Real-time speech-to-text translation over WebSocket. Streams Opus or raw PCM audio from client to server for live transcription and optional translation. Supports CLI and Python API.

asr2clip

Real-time speech-to-text clipboard tool with Silero VAD and local ASR support

taiwan-asr-toolkit

Production-grade Traditional Chinese / Taiwan Mandarin speech-to-text. Qwen3-ASR + MediaTek Breeze-ASR-25, hot-word injection, LLM polish, speaker diarization. RTF up to 1554x on RTX 5090, 56 TDD tests.

cohere-transcribe-arabic

Production-ready batch inference for Cohere’s Arabic/English ASR model, with optimized Silero VAD, multi-file GPU batching, subtitles, and optional word-level timestamps.

fast-silero-vad

High-throughput Silero VAD inference for CPU with multi-window ONNX execution and an optimized RFFT frontend.

kaiku

Prototype for a powerful voice-to-text pipeline for Linux & macOS. Flexibly chain denoiser, ASR, and AI post-processing components; Supports VAD and diarization; Local or cloud.

legionforge-convobox

A local, backend-agnostic voice frontend for CLI coding agents