MCP Video & Audio Text Extraction Server
A Model Context Protocol (MCP) server that enables text extraction from various video platforms and audio files, allowing compatible host applications (like Claude Desktop, Cursor) to access video content and perform text transcription.
What is it?
MCP Video & Audio Text Extraction Server is a Model Context Protocol (MCP) server that can download videos from various platforms, extract audio, and convert it to text. The server utilizes OpenAI's Whisper model for high-quality audio-to-text conversion.
How to use it?
Clone the repository and install dependencies
Ensure FFmpeg is installed
Run the server
Configure your MCP host application (like Claude Desktop) to use the server
Key Features
Support video downloads from multiple platforms including YouTube, Bilibili, TikTok, etc.
Extract audio content from videos
High-quality speech recognition using Whisper model
Multi-language text recognition support
Asynchronous processing for large files
Standardized MCP tools interface
Use Cases
Provide text transcription capabilities for applications that need to process video content
Batch process video content and extract text information
Create custom applications requiring audio/video text extraction functionality
Enable AI assistants to understand video content
FAQ
What are the system requirements to run the server?
> Requires Python 3.9+, FFmpeg, minimum 8GB RAM, GPU acceleration recommended
What should I know about first run?
> The system will automatically download the Whisper model file (approximately 1GB), which may take several minutes to tens of minutes
What audio formats are supported?
> Supports common audio formats including mp3, wav, m4a, etc.
This description maintains the core information from the original README while adopting a similar structure and style to the reference page. Would you like me to adjust or add anything to this description?