Supports downloading videos and extracting audio content for transcription using Whisper
Enables downloading videos and extracting audio content for transcription using Whisper
Supports downloading videos and extracting audio content for transcription using Whisper
Provides video downloading and audio extraction capabilities for transcription using Whisper
Uses OpenAI's Whisper model for high-quality multi-language audio-to-text transcription
Supports downloading audio content for transcription using Whisper
Enables downloading videos and extracting audio content for transcription using Whisper
Allows downloading videos and extracting audio content for transcription using Whisper
Allows downloading videos and extracting audio content for transcription using Whisper
MCP Video & Audio Text Extraction Server
An MCP server that provides text extraction capabilities from various video platforms and audio files. This server implements the Model Context Protocol (MCP) to provide standardized access to audio transcription services.
Supported Platforms
This service supports downloading videos and extracting audio from various platforms, including but not limited to:
YouTube
Bilibili
TikTok
Instagram
Twitter/X
Facebook
Vimeo
Dailymotion
SoundCloud
For a complete list of supported platforms, please visit yt-dlp supported sites.
Core Technology
This project utilizes OpenAI's Whisper model for audio-to-text processing through MCP tools. The server exposes four main tools:
Video download: Download videos from supported platforms
Audio download: Extract audio from videos on supported platforms
Video text extraction: Extract text from videos (download and transcribe)
Audio file text extraction: Extract text from audio files
MCP Integration
This server is built using the Model Context Protocol, which provides:
Standardized way to expose tools to LLMs
Secure access to video content and audio files
Integration with MCP clients like Claude Desktop
Features
High-quality speech recognition based on Whisper
Multi-language text recognition
Support for various audio formats (mp3, wav, m4a, etc.)
MCP-compliant tools interface
Asynchronous processing for large files
Tech Stack
Python 3.10+
Model Context Protocol (MCP) Python SDK
yt-dlp (YouTube video download)
openai-whisper (Core audio-to-text engine)
pydantic
System Requirements
FFmpeg (Required for audio processing)
Minimum 8GB RAM
Recommended GPU acceleration (NVIDIA GPU + CUDA)
Sufficient disk space (for model download and temporary files)
Important First Run Notice
Important: On first run, the system will automatically download the Whisper model file (approximately 1GB). This process may take several minutes to tens of minutes, depending on your network conditions. The model file will be cached locally and won't need to be downloaded again for subsequent runs.
Installation
Using uv (recommended)
When using uv no specific installation is needed. We will use uvx to directly run the video extraction server:
Install FFmpeg
FFmpeg is required for audio processing. You can install it through various methods:
Usage
Configure for Claude/Cursor
Add to your Claude/Cursor settings:
Available MCP Tools
Video download: Download videos from supported platforms
Audio download: Extract audio from videos on supported platforms
Video text extraction: Extract text from videos (download and transcribe)
Audio file text extraction: Extract text from audio files
Configuration
The service can be configured through environment variables:
Whisper Configuration
WHISPER_MODEL
: Whisper model size (tiny/base/small/medium/large), default: 'base'WHISPER_LANGUAGE
: Language setting for transcription, default: 'auto'
YouTube Download Configuration
YOUTUBE_FORMAT
: Video format for download, default: 'bestaudio'AUDIO_FORMAT
: Audio format for extraction, default: 'mp3'AUDIO_QUALITY
: Audio quality setting, default: '192'
Storage Configuration
TEMP_DIR
: Temporary file storage location, default: '/tmp/mcp-video'
Download Settings
DOWNLOAD_RETRIES
: Number of download retries, default: 10FRAGMENT_RETRIES
: Number of fragment download retries, default: 10SOCKET_TIMEOUT
: Socket timeout in seconds, default: 30
Performance Optimization Tips
GPU Acceleration:
Install CUDA and cuDNN
Ensure GPU version of PyTorch is installed
Model Size Adjustment:
tiny: Fastest but lower accuracy
base: Balanced speed and accuracy
large: Highest accuracy but requires more resources
Use SSD storage for temporary files to improve I/O performance
Notes
Whisper model (approximately 1GB) needs to be downloaded on first run
Ensure sufficient disk space for temporary audio files
Stable network connection required for YouTube video downloads
GPU recommended for faster audio processing
Processing long videos may take considerable time
MCP Integration Guide
This server can be used with any MCP-compatible client, such as:
Claude Desktop
Custom MCP clients
Other MCP-enabled applications
For more information about MCP, visit Model Context Protocol.
Documentation
For Chinese version of this documentation, please refer to README_zh.md
License
MIT
This server cannot be installed
hybrid server
The server is able to function both locally and remotely, depending on the configuration or use case.
An MCP server that downloads videos/extracts audio from various platforms like YouTube, Bilibili, and TikTok, then transcribes them to text using OpenAI's Whisper model.
Related MCP Servers
- AsecurityAlicenseAqualityA MCP server that enables transcription of audio files using OpenAI's Speech-to-Text API, with support for multiple languages and file saving options.Last updated -107MIT License
- -securityFlicense-qualitySimple MCP server that returns the transcription of a Youtube video using url and desired language.Last updated -
- -securityFlicense-qualityAn MCP server that generates structured notes from Bilibili videos by automatically downloading audio, transcribing with Whisper, and processing through LLM.Last updated -6
- -securityFlicense-qualityA powerful speech-to-text MCP server that supports multiple audio formats and recognition engines including remote APIs (Bailian, OpenAI Whisper, iFLYTEK), Google Speech Recognition, and CMU Sphinx.Last updated -