OptionalmodelWhere the model comes from — self-host it by mirroring the repo: the file is fetched from
<modelHost>/onnx-community/silero-vad/resolve/main/onnx/model_quantized.onnx.
Default https://huggingface.co.
OptionalruntimeWhere the runtime's .wasm files are served (default: next to its JS module).
Options of createSpeechDetector.