This project provides a framework for creating an AI-powered Virtual YouTuber (Vtuber) that interacts with a live YouTube audience. It leverages OpenAI's GPT-3 for chat response generation and ElevenLabs for text-to-speech, enabling a dynamic and interactive streaming experience. The target audience includes developers and content creators interested in AI-driven entertainment and virtual personas.

How It Works

The system ingests YouTube chat messages, processes them through OpenAI's GPT-3 to generate contextually relevant responses, and then vocalizes these responses using ElevenLabs' text-to-speech engine. This creates an automated, conversational AI that can engage with viewers in real-time.

Quick Start & Requirements

• Install dependencies: pip install -r requirements.txt
• Requires ffmpeg to be installed.
• API keys for ElevenLabs (EL_key) and OpenAI (OAI_key) are needed, configured in config.json.
• Run with python run.py -id STREAMID.
• Official docs: Not explicitly linked, but voice IDs can be found via a provided link.

Highlighted Details

• Integrates with YouTube chat for real-time interaction.
• Utilizes GPT-3 for natural language understanding and response generation.
• Employs ElevenLabs for high-quality speech synthesis.
• Supports mouth animation via VTS plugin and VB-Audio cable integration.

Maintenance & Community

The project was created for fun and a YouTube video; significant updates or improvements are not expected after the initial checklist completion.

Licensing & Compatibility

Licensed under the MIT license, permitting commercial use and integration with closed-source projects.

Limitations & Caveats

The code quality and reliability are noted as potentially questionable due to its experimental nature. The project is not actively maintained for further improvements.