# snakers4/silero-vad

Silero VAD: pre-trained enterprise-grade Voice Activity Detector

Repository: https://github.com/snakers4/silero-vad
Canonical: https://ross.abutalabs.com/products/silero-vad
Language: Python
License: MIT
License Family: permissive
Topics: voice-detection, voice-recognition, voice-commands, pytorch, onnx, voice-activity-detection, voice-control, onnx-runtime, onnxruntime, speech, speech-processing, vad
Last push: 2026-08-24T06:47:25+00:00

## Health v2 (maintenance only)
Score: 86/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 99, release rhythm 60, longevity 100
- inputs: {"age_days": 2109, "days_push": 9, "days_rel": 190, "gap_med": 70, "n_releases_24m": 6}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 10061, forks 833 (observed 2026-08-28T04:10:39.383682+00:00)

## What it is
Silero VAD is a pre-trained, enterprise-grade Voice Activity Detector model available via PyPI, runnable with PyTorch or ONNX Runtime. It detects speech segments in audio streams, including in real time, with a lightweight footprint.

## Use cases
- detect speech segments in audio recordings
- run real-time voice activity detection in a streaming pipeline
- segment audio before feeding a speech-to-text model
- filter out silence and noise from voice recordings
- add voice activation to a voice assistant or command app
- run VAD on edge devices via onnxruntime

## When to choose
- you need a fast, lightweight, pre-trained VAD without training your own model
- you need real-time speech detection with low latency
- you want ONNX support for deployment across architectures
- you're building ASR pipelines and need audio segmentation

## When to avoid
- you need full speech recognition or transcription, not just speech detection
- you need speaker identification or diarization
- you can't use PyTorch or ONNX Runtime as dependencies

## Facets
- artifact type: library
- maturity: stable
- function: speech-recognition, audio-processing, machine-learning, llm-inference
- domain: speech-processing, machine-learning
- platform: python, cross-platform
- tags: voice-activity-detection, vad, onnx, pytorch, pre-trained-model, real-time-audio, audio, natural-language-processing, gpu

## Member repositories
- snakers4/silero-vad (main) score 86

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:10:39.383682+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T17:19:53.594539+00:00, confidence not recorded.
  - readme: https://github.com/snakers4/silero-vad (fetched 2026-08-28T04:10:39.383682+00:00, sha 07b4d90f7569)
  - registry_pypi: https://pypi.org/pypi/silero-vad/json (fetched 2026-08-29T08:19:37.609927+00:00, sha a282780616ea)
- Data as of 2026-08-30T08:39:29.467469+00:00.