Top AI Repos — open-source AI, indexed and scored
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
Voice Activity Detection based on Deep Learning & TensorFlow
| Date | Stars |
|---|---|
| 2026-07-24 | 373 |
| 2026-07-25 | 373 |
| 2026-07-28 | 373 |
| 2026-07-30 | 373 |
| 2026-07-31 | 374 |
| 2026-08-06 | 374 |
Today
— stars today
This week
+1 stars this week
This month
— stars this month
Momentum
1.0
growth rate 0.27%/day
# Voice Activity Detection project
<p align="center">
<a href="https://github.com/filippogiruzzi/voice_activity_detection/actions/workflows/ci.yml" alt="CI">
<img src="https://github.com/filippogiruzzi/voice_activity_detection/actions/workflows/ci.yml/badge.svg" /></a>
<a href="https://github.com/filippogiruzzi/voice_activity_detection/actions/workflows/cd.yml" alt="CD">
<img src="https://github.com/filippogiruzzi/voice_activity_detection/actions/workflows/cd.yml/badge.svg" /></a>
</p>
<p align="center">
<a href="https://github.com/filippogiruzzi/voice_activity_detection"><img src="https://img.shields.io/github/stars/filippogiruzzi/voice_activity_detection?logo=github" alt="GitHub stars"></a>
<a href="https://github.com/filippogiruzzi/voice_activity_detection"><img src="https://img.shields.io/github/forks/filippogiruzzi/voice_activity_detection?logo=github" alt="GitHub forks"></a>
<a href="https://hub.docker.com/repository/docker/filippogrz/vad"><img src="https://img.shields.io/docker/pulls/filippogrz/vad?logo=docker" alt="Docker Pulls"></a>
</p>
<center>Keywords: Python, PyTorch, Deep Learning, Audio, Time Series classification, uv, Docker</center>
## Table of contents
1. [ Installation ](#1-installation)
2. [ Introduction ](#2-introduction)
3. [ Project structure ](#3-project-structure)
4. [ Dataset ](#4-dataset)
5. [ Project usage ](#5-project-usage)
6. [ Testing ](#6-testing)
7. [ Continuous integration & delivery ](#7-continuous-integration--delivery)
8. [ Contributing ](#8-contributing)
9. [ Todo ](#9-todo)
10. [ License ](#10-license)
11. [ Resources ](#11-resources)
## 1. Installation
This project uses:
* Python 3.11+
* PyTorch 2.2+
* [uv](https://docs.astral.sh/uv/) package manager
```bash
git clone https://github.com/filippogiruzzi/voice_activity_detection.git
cd voice_activity_detection/
```
### 1.1 Install with uv (recommended)
```bash
# Install uv if you don't have it
curl -LsSf https://astral.sh/uv/install.sh | sh
# Install all dependencies
uv sync
```
### 1.2 Development setup
```bash
# Install with dev dependencies (ruff, pytest, etc.)
uv sync
# Run linters
make lint
# Run tests
make test
# Auto-format code
make format
```
### 1.3 Docker installation
Build and run the CPU Docker image:
```bash
make build
make local-nobuild
```
For GPU support (requires NVIDIA Docker runtime):
```bash
make build-gpu
```
## 2. Introduction
### 2.1 Goal
The purpose of this project is to design and implement a real-time Voice Activity Detection algorithm based on Deep Learning.
The designed solution is based on a simple pipeline with MFCC feature extraction and a small 1D-ResNet model (PyTorch) that classifies whether an audio signal is speech or noise.
### 2.2 Results
| Model | Train acc. | Val acc. | Test acc. |
| :---: |:---:| :---:| :---: |
| 1D-Resnet | 99 % | 98 % | 97 % |
Raw and post-processed inference results on a test audio signal are shown below.
### 2.3 Model & features
Each audio window of `SEQ_LEN = 1024` samples (16 kHz) is converted into a
**16 × 65** feature tensor stacking:
* 5 MFCC coefficients,
* 5 MFCC deltas (1st order),
* 5 MFCC delta-deltas (2nd order),
* 1 RMS energy.
These features feed a configurable 1D-ResNet (`vad.model.Resnet1D`):
stacked residual blocks (3 × `Conv1d → BatchNorm1d` with a 1×1 shortcut) →
global average pooling → a fully connected head producing a single speech logit.
The architecture is fully described by the `ModelConfig` dataclass, so the same
configuration must be used at training, export, and inference time.
## 3. Project structure
The core code lives flat inside `vad/`:
* `vad/model.py`: the `Resnet1D` model architecture and its `ModelConfig` dataclass
* `vad/data.py`: feature extraction, dataset building & the PyTorch DataLoader
* `vad/train.py`: training loop & model export (state Excerpt of 8,568 characters
Read on GitHubWould you bet a product on this? Bounded 0–100 and slow moving.
matched fp:3767f3f0df027655, topic:deep-learning, topic:pytorch
matched fp:3767f3f0df027655, topic:speech-recognition, readme:real-time voice