Whisper Speech Recognition MCP Server

Whisper Speech Recognition MCP Server

A high-performance speech recognition MCP server based on Faster Whisper, providing efficient audio transcription capabilities.

BigUncle

Research & Data
Visit Server

README

Whisper Speech Recognition MCP Server


中文文档

A high-performance speech recognition MCP server based on Faster Whisper, providing efficient audio transcription capabilities.

Features

  • Integrated with Faster Whisper for efficient speech recognition
  • Batch processing acceleration for improved transcription speed
  • Automatic CUDA acceleration (if available)
  • Support for multiple model sizes (tiny to large-v3)
  • Output formats include VTT subtitles, SRT, and JSON
  • Support for batch transcription of audio files in a folder
  • Model instance caching to avoid repeated loading
  • Dynamic batch size adjustment based on GPU memory

Installation

Dependencies

  • Python 3.10+
  • faster-whisper>=0.9.0
  • torch==2.6.0+cu126
  • torchaudio==2.6.0+cu126
  • mcp[cli]>=1.2.0

Installation Steps

  1. Clone or download this repository
  2. Create and activate a virtual environment (recommended)
  3. Install dependencies:
pip install -r requirements.txt

PyTorch Installation Guide

Install the appropriate version of PyTorch based on your CUDA version:

  • CUDA 12.6:

    pip install torch==2.6.0 torchvision==0.21.0 torchaudio==2.6.0 --index-url https://download.pytorch.org/whl/cu126
    
  • CUDA 12.1:

    pip install torch==2.5.1 torchvision==0.20.1 torchaudio==2.5.1 --index-url https://download.pytorch.org/whl/cu121
    
  • CPU version:

    pip install torch==2.6.0 torchvision==0.21.0 torchaudio==2.6.0 --index-url https://download.pytorch.org/whl/cpu
    

You can check your CUDA version with nvcc --version or nvidia-smi.

Usage

Starting the Server

On Windows, simply run start_server.bat.

On other platforms, run:

python whisper_server.py

Configuring Claude Desktop

  1. Open the Claude Desktop configuration file:

    • Windows: %APPDATA%\Claude\claude_desktop_config.json
    • macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
  2. Add the Whisper server configuration:

{
  "mcpServers": {
    "whisper": {
      "command": "python",
      "args": ["D:/path/to/whisper_server.py"],
      "env": {}
    }
  }
}
  1. Restart Claude Desktop

Available Tools

The server provides the following tools:

  1. get_model_info - Get information about available Whisper models
  2. transcribe - Transcribe a single audio file
  3. batch_transcribe - Batch transcribe audio files in a folder

Performance Optimization Tips

  • Using CUDA acceleration significantly improves transcription speed
  • Batch processing mode is more efficient for large numbers of short audio files
  • Batch size is automatically adjusted based on GPU memory size
  • Using VAD (Voice Activity Detection) filtering improves accuracy for long audio
  • Specifying the correct language can improve transcription quality

Local Testing Methods

  1. Use MCP Inspector for quick testing:
mcp dev whisper_server.py
  1. Use Claude Desktop for integration testing

  2. Use command line direct invocation (requires mcp[cli]):

mcp run whisper_server.py

Error Handling

The server implements the following error handling mechanisms:

  • Audio file existence check
  • Model loading failure handling
  • Transcription process exception catching
  • GPU memory management
  • Batch processing parameter adaptive adjustment

Project Structure

  • whisper_server.py: Main server code
  • model_manager.py: Whisper model loading and caching
  • audio_processor.py: Audio file validation and preprocessing
  • formatters.py: Output formatting (VTT, SRT, JSON)
  • transcriber.py: Core transcription logic
  • start_server.bat: Windows startup script

License

MIT

Acknowledgements

This project was developed with the assistance of these amazing AI tools and models:

Special thanks to these incredible tools and the teams behind them.

Recommended Servers

Crypto Price & Market Analysis MCP Server

Crypto Price & Market Analysis MCP Server

A Model Context Protocol (MCP) server that provides comprehensive cryptocurrency analysis using the CoinCap API. This server offers real-time price data, market analysis, and historical trends through an easy-to-use interface.

Featured
TypeScript
MCP PubMed Search

MCP PubMed Search

Server to search PubMed (PubMed is a free, online database that allows users to search for biomedical and life sciences literature). I have created on a day MCP came out but was on vacation, I saw someone post similar server in your DB, but figured to post mine.

Featured
Python
dbt Semantic Layer MCP Server

dbt Semantic Layer MCP Server

A server that enables querying the dbt Semantic Layer through natural language conversations with Claude Desktop and other AI assistants, allowing users to discover metrics, create queries, analyze data, and visualize results.

Featured
TypeScript
mixpanel

mixpanel

Connect to your Mixpanel data. Query events, retention, and funnel data from Mixpanel analytics.

Featured
TypeScript
Sequential Thinking MCP Server

Sequential Thinking MCP Server

This server facilitates structured problem-solving by breaking down complex issues into sequential steps, supporting revisions, and enabling multiple solution paths through full MCP integration.

Featured
Python
Nefino MCP Server

Nefino MCP Server

Provides large language models with access to news and information about renewable energy projects in Germany, allowing filtering by location, topic (solar, wind, hydrogen), and date range.

Official
Python
Vectorize

Vectorize

Vectorize MCP server for advanced retrieval, Private Deep Research, Anything-to-Markdown file extraction and text chunking.

Official
JavaScript
Mathematica Documentation MCP server

Mathematica Documentation MCP server

A server that provides access to Mathematica documentation through FastMCP, enabling users to retrieve function documentation and list package symbols from Wolfram Mathematica.

Local
Python
kb-mcp-server

kb-mcp-server

An MCP server aimed to be portable, local, easy and convenient to support semantic/graph based retrieval of txtai "all in one" embeddings database. Any txtai embeddings db in tar.gz form can be loaded

Local
Python
Research MCP Server

Research MCP Server

The server functions as an MCP server to interact with Notion for retrieving and creating survey data, integrating with the Claude Desktop Client for conducting and reviewing surveys.

Local
Python