Skip to content

Latest commit

 

History

History
79 lines (56 loc) · 3.29 KB

snakers4_silero-vad_vad.md

File metadata and controls

79 lines (56 loc) · 3.29 KB
Raw
layout background-class body-class category title summary image author tags github-link github-id featured_image_1 featured_image_2 accelerator demo-model-link
hub_detail
hub-background
hub
researchers
Silero Voice Activity Detector
Pre-trained Voice Activity Detector
silero_logo.jpg
Silero AI Team
audio
scriptable
snakers4/silero-vad
silero_vad_performance.png
no-image
cuda-optional
 
# this assumes that you have a proper version of PyTorch already installed
pip install -q torchaudio soundfile
import torch
torch.set_num_threads(1)
from pprint import pprint

model, utils = torch.hub.load(repo_or_dir='snakers4/silero-vad',
                              model='silero_vad',
                              force_reload=True)

(get_speech_ts,
 _, _, read_audio,
 _, _, _) = utils

files_dir = torch.hub.get_dir() + '/snakers4_silero-vad_master/files'

wav = read_audio(f'{files_dir}/en.wav')
speech_timestamps = get_speech_ts(wav, model,
                                  num_steps=4)
pprint(speech_timestamps)

Model Description

Silero VAD: pre-trained enterprise-grade Voice Activity Detector (VAD), Number Detector and Language Classifier. Enterprise-grade Speech Products made refreshingly simple (see our STT models). Each model is published separately.

Currently, there are hardly any high quality / modern / free / public voice activity detectors except for WebRTC Voice Activity Detector (link). WebRTC though starts to show its age and it suffers from many false positives.

Also in some cases it is crucial to be able to anonymize large-scale spoken corpora (i.e. remove personal data). Typically personal data is considered to be private / sensitive if it contains (i) a name (ii) some private ID. Name recognition is a highly subjective matter and it depends on locale and business case, but Voice Activity and Number Detection are quite general tasks.

(!!!) Important Notice (!!!) - the models are intended to run on CPU only and were optimized for performance on 1 CPU thread. Note that the model is quantized.

Supported Languages

As of this page update, the following languages are supported:

  • Russian
  • English
  • German
  • Spanish

Please note that in theory the VAD should also work fine with similar / related languages (e.g. Germanic, Slavic or Romance languages). To see the always up-to-date language list, please visit our repo.

Additional Examples and Benchmarks

For additional examples and other model formats please visit this link and please refer to the extensive examples in the Colab format (including the streaming examples).

References

VAD model architectures are based on similar STT architectures.