GitHub - snakers4/silero-vad: Silero VAD: pre-trained enterprise-grade Voice Activity Detector

Stellar accuracy Silero VAD has excellent results on speech detection tasks. Fast One audio chunk (30+ ms) takes less than 1ms to be processed on a single CPU thread. Using batching or GPU can also improve performance considerably. Under certain conditions ONNX may even run up to 4-5x faster. Lightweight JIT model is around one megabyte in size. General Silero VAD was trained on huge corpora that

[stealthinu]

軽そうなVAD。deeplearning使ってるがモデルは小さそうでONNXモデルの提供あってCPUでの動作も考えられている。