Practical overview
When it fits
Consider it when evaluating a Python voice-activity-detection project for a speech workflow and you are prepared to inspect the repository for implementation and usage details.
Who it is for
- Developers exploring voice-activity detection as a component of a speech-processing pipeline.
- Python developers assessing a GPL-3.0 project for speech-related work.
Documented uses
- Explore a deep-learning approach to identifying voice activity in audio.
- Assess a potential voice-activity-detection component for a larger speech pipeline.
Requirements & limits
- The project description names TensorFlow, while PyTorch appears only in repository topics; the implementation details are not described.
- The repository is licensed under GPL-3.0, a relevant consideration when assessing reuse.
- Installation and inference instructions are not available in the provided project description.
Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.
Install the project with uv
Set up the repository and install its dependencies for local development.
Before you start
- Python 3.11 or later
- The uv package manager
Steps
Clone the repository: the command below.
git clone https://github.com/filippogiruzzi/voice_activity_detection.gitChange into the repository directory:
cd voice_activity_detection/.Install dependencies with the command below.
uv syncOptionally, run the command below, the command below, or the command below for the documented development checks and formatting.
make lintmake testmake format
Expected outcome
The repository is checked out and its dependencies are installed; the documented development commands are available to run.
Guide limitations
- This setup procedure does not create a dataset, train a model, or run inference.
- The documented dataset stage requires LibriSpeech; see the project usage instructions for dataset creation.
Create a processed dataset from LibriSpeech audio
Convert raw LibriSpeech audio into the processed dataset files used by the training stage.
Before you start
- Python 3.11 or later, PyTorch 2.2 or later, and uv
- LibriSpeech audio downloaded and available in a directory
Steps
Install the repository dependencies with the command below from the repository directory.
uv syncRun the command below, replacing the path with the location of the downloaded audio.
uv run vad-data --data-dir /path/to/LibriSpeech/Optionally add
--max-files Nto process only a limited number of files per split.
Expected outcome
Processed `.pt` files are saved under the specified data directory in `dataset/{train,val,test}/`, ready for the documented training stage.
Guide limitations
- The README says to download LibriSpeech but does not include the dataset download procedure in the available text.
- This recipe creates data only; model training and inference are separate stages.
Sources & review
Source record details
- Source availability
- Active
- Sources checked
- 8 October 2026
- Original website title
- filippogiruzzi/voice_activity_detection · Official homepage title · retained in its source language