Polyglot is a cross-platform AI-powered language practice application designed for users seeking to improve their spoken fluency in multiple languages. It leverages large language models like ChatGPT and Azure AI for conversational practice, offering features such as AI-generated speech, text translation, and customizable dialogue scenarios.

How It Works

The application utilizes ChatGPT for intelligent dialogue and Azure TTS for speech synthesis, providing a natural conversational experience. Users can practice speaking with AI personas representing different nationalities, with the ability to customize languages, AI characters, and dialogue scenarios. Speech recognition is integrated, allowing users to speak and receive feedback.

Quick Start & Requirements

• Install via pnpm install and run with pnpm dev.
• Requires Node.js and pnpm.
• Cloning the repository is necessary.
• Official releases are available on GitHub Releases.

Highlighted Details

• Supports multiple languages with customizability for languages and AI characters.
• Integrates Azure TTS for speech synthesis and ChatGPT for dialogue.
• Features a web version and desktop applications for Mac, Windows, and Linux.
• Includes speech recognition with hotkey support and self-playback of user speech.

Maintenance & Community

• The project is maintained by liou666.
• Community support channels are not explicitly mentioned in the README.

Licensing & Compatibility

• Licensed under GNU General Public License v3.0 (GPL-3.0).
• GPL-3.0 is a strong copyleft license, requiring derivative works to also be open-sourced under the same license.

Limitations & Caveats

The web version has limited compatibility with mobile devices. macOS users may encounter Gatekeeper security warnings requiring manual bypass or whitelisting. The project is primarily focused on Azure TTS and OpenAI/Azure OpenAI APIs, with potential future integration of other services.