Back: Catalog

Silero VAD

Silero VAD is a pre-trained audio activity detection tool based on PyTorch and ONNX. Supporting 8000 Hz and 16,000 Hz sampling speeds, the model can process audio stacks longer than 30 milliseconds in a single CPU thread in less than 1 millisecond. It offers a lightweight structure with a JIT model of about two megabytes in size and is trained in thousands of different languages.

SI

Silero VAD

100

snakers4 · AI Call Agents & Telephony

STARS10.6K
FORKS879
COMMITS556
CONTRIBUTORS52
VERSIONv6.2.1
UPDATED9 days ago
Python / OnnxBarge-InDil BağımsızReal-TimeMIT

Common commercial products in this category: Vapi, Bland AI, Retell AI

What can it do?
  • Using Python and pips, you can integrate speech in audio files into your project to search for time.
  • By calling the model via the PyTorch Hub, you can process audio data and get speech times in seconds.
  • Using ONNX Runtime, you can customize and execute audio I/O operations in different system architectures.
User reviews
  • • NeutralGitHub Issue

    [CHANGELOG] v6.2 just released!

    “Just a handy issue to be notified of latest changes and micro-releases (we will mostly changing the models)”

  • • NeutralGitHub Issue

    [FORMATS] Most important export formats?

    “As everybody knows, the VAD now exists as an ONNX model (`onnx-runtime`) and as a `jit` (PyTorch TorchScript) model. Can you please shorlist the remaining most important export formats, i.e. TF, OpenVINO, etc? We need some data / votes to make a shortlist. To make it simpler - I…”

GitHub

My note

Alternative pages

Vapi alternativesBland AI alternativesRetell AI alternatives

More in: AI Call Agents & Telephony