Babelfish.ai provides real-time, in-browser speech-to-text transcription and translation for 200 languages. It targets users needing live multilingual communication, leveraging Hugging Face's Transformer.js and Supabase Realtime for local, browser-based operation.

How It Works

The application utilizes OpenAI's Whisper model via transcriptionWorker.js for local speech-to-text. Translated output is broadcast to subscribed clients using Supabase Realtime. Meta's NLLB-200 model handles translation to over 200 languages, also processed locally within the browser.

Quick Start & Requirements

• Install: npm run dev
• Prerequisites: Supabase credentials (set in .env.local).
• Deployment: Push to GitHub main branch for GitHub Pages deployment.
• Demo: Watch a demo / tutorial

Highlighted Details

• Real-time, in-browser transcription using OpenAI Whisper.
• Live translation to 200 languages via Meta's NLLB-200.
• Broadcasts transcriptions and translations using Supabase Realtime.

Maintenance & Community

• Project is part of the Supabase community.
• No specific community links (Discord/Slack) or roadmap are provided in the README.

Licensing & Compatibility

• License: Not specified in the README.
• Compatibility: Designed for browser-based execution.

Limitations & Caveats

The project's license is not specified, which may impact commercial use. The README does not detail performance benchmarks or specific browser compatibility limitations.