vad crdnn libriparty EU AI Act Compliance Profile

Voice Activity Detection with a (small) CRDNN model trained on Libriparty

This repository provides all the necessary tools to perform voice activity detection with SpeechBrain using a model pretrained on Libriparty.

The pre-trained system can process short and long speech recordings and outputs the segments where speech activity is detected. The output of the system looks like this:

The system expects input recordings sampled at 16kHz (single channel). If your signal has a different sample rate, resample it (e.g., using torchaudio or sox) before using the interface.

For a better experience, we encourage you to learn more about SpeechBrain.

Results

The model performance on the LibriParty test set is:

| Release | hyperparams file | Test Precision | Test Recall | Test F-Score | Model link | GPUs | |:-------------:|:---------------------------:| -----:| -----:| --------:| :-----------:| :-----------:| | 2021-09-09 | train.yaml | 0.9518 | 0.9437 | 0.9477 | Model | 1xV100 16GB

Pipeline description

This system is composed of a CRDNN that outputs posteriors probabilities with a value close to one for speech frames and close to zero for non-speech segments. A threshold is applied on top of the posteriors to detect candidate speech boundaries.

Depending on the active options, these boundaries can be post-processed (e.g, merging close segments, removing

Your obligations as a vad crdnn libriparty deployer

Track your progress

Sign up to track obligations, receive deadline reminders, and generate audit packages.

Tags

Languages

Library