sherpa-onnx-silero-vad-ax650

Silero VAD (Voice Activity Detection) model converted to AX650 AXMODEL for on-device inference via sherpa-onnx.

Model Details

  • Original Model: k2-fsa/sherpa-onnx-silero-vad
  • Architecture: CNN-based VAD
  • Task: Voice Activity Detection
  • Sample rate: 16000 Hz
  • Window size: 512 samples
  • Context size: 64 samples
  • Hidden size: 128
  • Target Chip: AX650 (NPU3)
  • Quantization: U8

Files

File Size Description
silero_vad.axmodel 358 KB AX650 VAD model

Usage with sherpa-onnx

./sherpa-onnx-vad \
  --silero-vad-model=silero_vad.axmodel \
  --vad-provider=axera \
  input.wav output.wav

Performance

Metric Value
AXMODEL size 358 KB
Verified on AX650N DEMO Board

Conversion Details

  • Pulsar2 Version: 6.0
  • Calibration: MinMax with 10 samples
  • Quantization: U8

License

Apache 2.0

Downloads last month

-

Downloads are not tracked for this model. How to track
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Collection including inoryQwQ/sherpa-onnx-silero-vad-ax650