Desktop Audio Translator

A desktop audio translator with offline Whisper speech recognition, Helsinki-NLP translation, and a simple GUI with language switching.

Desktop Audio Translator
🎤 Source: مرحبا كيف حالك اليوم؟
🔤 Target: Hello, how are you today?
─────────────────────────────
🎤 Source: أهلا وسهلا بك
🔤 Target: Welcome

Powerful Features

Real-time Audio Capture

Capture audio from any system device with smart device selection and persistent preferences.

  • Desktop audio capture
  • Microphone input support
  • Device switching with Ctrl+D

Advanced Speech Recognition

Offline speech recognition using Whisper models (runs locally after download).

  • Offline Whisper ASR
  • GPU acceleration (CUDA)
  • Configurable model size

Instant Translation

Translate from a selectable source language to a selectable target language.

  • Helsinki-NLP models
  • Download per language pair
  • Offline after download

Session Management

Automatic transcript saving with organized output and configuration persistence.

  • Timestamped transcripts
  • Automatic file organization
  • Session metadata

GUI + CLI

Use the simple GUI for language selection, or the CLI for quick control.

  • GUI: device + language picker
  • GUI: download models button
  • CLI: Ctrl+D / Ctrl+C shortcuts

Easy Setup

Simple installation process with comprehensive documentation and troubleshooting.

  • One-command installation
  • Cross-platform support
  • Detailed documentation

Get Started

Quick Installation

Terminal
# Clone or download the repository
git clone https://github.com/tanzir71/arabic-desktop-audio-translator.git
cd arabic-desktop-audio-translator

# Install dependencies
pip install -r requirements.txt

# Run the GUI
python gui.py

# Or run CLI mode
python main.py

System Requirements

  • Python: 3.8 or higher
  • OS: Windows
  • RAM: 4GB minimum (8GB recommended)
  • Storage: 2GB free space for models
  • Internet: Only required for first-time model downloads