FP16 ONNX export of Silero VAD, an open-weight voice activity detection model. The immediate upstream source is onnx-community/silero-vad; the original model is from the Silero team (snakers4/silero-vad on GitHub). This is not a newly trained model.

Source

Field	Value
Upstream model	onnx-community/silero-vad
Upstream source revision	`e71cae966052b992a7eca6b17738916ce0eca4ec`
Export tool/script	Mirrored from onnx-community (no re-export)
Quantization recipe	FP16 ONNX mirrored from `onnx/model_fp16.onnx`
Original model	snakers4/silero-vad (GitHub)

Files

File	Size	Description
`onnx/model_fp16.onnx`	~1 MB	FP16 ONNX model

Intended Use

Voice activity detection for local inference via ONNX Runtime. VAD identifies segments of audio that contain speech versus silence or background noise. The FP16 variant targets GPU execution; the INT8 variant is in tonythethompson/silero-vad-int8-onnx. Silero VAD is language-agnostic and works across a wide range of languages.

Runtime Notes

Designed for ONNX Runtime compatible runtimes.
Expected input: 16 kHz mono audio (Silero VAD requirement).
Execution: GPU (CUDA) preferred for FP16. Validate on the target execution provider before production use.

Precision and Packaging

Export tooling, precision, and quantization are recorded in the Source table above. This packaging mirror does not publish independent parity benchmarks; validate on your target execution provider before production use.

Limitations

Requires 16 kHz mono audio input; other sample rates must be resampled.
FP16 precision may cause degraded detection on hardware that upcasts silently.
No repository-specific VAD quality benchmark is documented here.
Performance on very noisy or music-heavy audio is not evaluated here.

License

MIT — inherited from onnx-community/silero-vad and the original Silero VAD model. This packaging repo adds no new license terms.

tonythethompson/Silero-VAD-FP16-ONNX

Silero VAD — FP16 ONNX