This project generates "brainrot" style videos, transforming text into video content featuring popular personalities. It targets users interested in creating engaging, personality-driven educational or entertainment content, offering a novel way to learn about topics through synthesized speech and visuals.

How It Works

The system leverages Docker for environment management and relies on multiple API integrations for its core functionality. It uses Groq for fast LLM inference to process input text and generate scripts, OpenAI for additional processing, and Speechify for high-quality voice cloning of various personalities. The output is a video file, with a "studio mode" allowing real-time editing of video composition code.

Quick Start & Requirements

• Install Docker.
• Create a .env file with API keys for Speechify, Groq, and OpenAI, along with specific voice IDs for personalities from Speechify.
• Run ./scripts/build.sh to build the Docker image (approx. 5-15 minutes, 5.5GB image size).
• Run bun install in the ./generate directory.
• Execute ./scripts/start.sh for regular video generation or MODE=studio ./scripts/start.sh for studio mode.
• API key acquisition links: Speechify (https://speechify.com/text-to-speech-api/), OpenAI (https://platform.openai.com/api-keys), Groq (https://console.groq.com/keys).

Highlighted Details

• Utilizes Groq for potentially faster LLM processing.
• Supports voice cloning for multiple distinct personalities via Speechify.
• Offers a "studio mode" for real-time video code editing and preview.
• Video generation process can take 10-20 minutes.

Maintenance & Community

No specific information on contributors, sponsorships, or community channels is provided in the README.

Licensing & Compatibility

The README does not specify a license.

Limitations & Caveats

Requires significant local storage (up to 12.6GB). The "rap mode" is mentioned but not documented. The project appears to be in active development, with potential for undocumented features or changes.