Skip to content

Latest commit

 

History

History
101 lines (63 loc) · 3.21 KB

README.md

File metadata and controls

101 lines (63 loc) · 3.21 KB
Raw

Real-Time Audio Translator

Overview

The Real-Time Audio Translator is a Python application that captures audio from selected input devices, transcribes the audio using the Whisper model, and then translates the transcription into a specified target language in real-time. The terminal interface is managed using ncurses, providing a user-friendly display of both the transcription and the translation.

Features

  • Real-Time Audio Capture: Capture audio from selected input devices.
  • Transcription: Transcribe audio using the Whisper model.
  • Translation: Translate the transcribed text using the OPUS-MT models.
  • Terminal Interface: Display transcriptions and translations in a user-friendly terminal interface using ncurses.
  • Multi-language Support: Supports multiple languages for both transcription and translation.

Installation

Prerequisites

  • Python 3.10 or higher
  • A working installation of PyAudio
  • Other Python dependencies listed in requirements.txt

Steps

  1. Clone the repository: 
    git clone <repository_url>
cd realtime-translator
  2. Create a virtual environment and activate it: 
    python3 -m venv .venv
source .venv/bin/activate
  3. Install the required Python packages: 
    pip install -r requirements.txt
  4. Run the application: 
    python main.py

Usage

When you run the application, you will be prompted to:

  1. Select Audio Input Devices: Choose the audio input devices you want to use.
  2. Specify Languages: Choose the source language for transcription and the destination language for translation.

The terminal interface will display the real-time transcription and translation.

File Descriptions

audio_config.py

Handles audio source selection and language configuration. It prompts the user to select audio input devices and specify the source and destination languages.

audio_processing.py

Manages the audio capture, transcription, and translation processes. It also handles the queuing of audio data to ensure only one process runs at a time.

main.py

The entry point of the application. It manages the initialization of the terminal interface, the setup of models, and the main loop for processing audio.

sources.py

Run this directly to view your audio interfaces.

terminal_interface.py

Manages the terminal interface using ncurses. It displays transcriptions and translations in real-time and handles cleanup of terminal resources.

utils.py

Provides utility functions such as error message formatting.

.editorconfig

Configures coding styles such as indentation, charset, end-of-line characters, etc.

.gitignore

Specifies files and directories to be ignored by Git.

requirements.txt

Lists the Python packages required to run the application.

Contributing

Contributions are welcome! Please fork this repository and submit a pull request with your improvements.

License

This project is licensed under the MIT License.

Acknowledgements

This project leverages the Whisper model for transcription and the OPUS-MT models for translation. Special thanks to the Hugging Face community for their contributions to these models.