simple-asr-mcp
Minimal MCP server for local speech recognition using faster-whisper. Runs on CPU, no cloud required.
README
simple-asr-mcp
Minimal MCP server for local speech recognition using faster-whisper. Runs on CPU, no cloud required.
Installation
pipx install simple-asr-mcp
Or with pip:
pip install simple-asr-mcp
Requirements
- Python 3.10+
- ffmpeg (required by faster-whisper for audio decoding)
On Windows, install ffmpeg via winget:
winget install Gyan.FFmpeg
Quick Setup (for Claude Code agents)
Run these two commands — no user interaction required:
pipx install simple-asr-mcp
claude mcp add asr --scope user -- simple-asr-mcp
Then ask the user to restart Claude Code or run /mcp to reconnect.
Usage
CLI
# Transcribe an audio file
simple-asr-mcp transcribe recording.wav
# Specify language and model
simple-asr-mcp transcribe recording.wav --language ru --model medium
# List available models
simple-asr-mcp models
MCP Server (Claude Code)
MCP tools available after setup:
- transcribe_file — transcribe any audio file by path
- list_models — see available Whisper models
Configuration
Environment variables:
| Variable | Default | Description |
|---|---|---|
WHISPER_MODEL |
small |
Default Whisper model |
WHISPER_DEVICE |
cpu |
Device: cpu, cuda, or auto |
WHISPER_COMPUTE_TYPE |
int8 |
Quantization type |
Example with custom config:
claude mcp add asr --scope user -e WHISPER_MODEL=medium -e WHISPER_DEVICE=cuda -- simple-asr-mcp
Available Models
| Model | Size | RAM (est.) |
|---|---|---|
| tiny | 75 MB | ~1 GB |
| base | 142 MB | ~1 GB |
| small | 466 MB | ~2 GB |
| medium | 1.5 GB | ~5 GB |
| large-v3 | 3.1 GB | ~10 GB |
The model is downloaded automatically on first use and cached locally. It stays in memory until the MCP server process exits.
Supported Audio Formats
Any format supported by ffmpeg: wav, mp3, flac, ogg, m4a, wma, etc.
Links
License
MIT
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.