Gladia MCP

Features

Audio transcription with speaker diarization
Real-time speech-to-text
Audio intelligence capabilities:
- Translation
- Summarization
- Named Entity Recognition
- Sentiment Analysis
- Content Moderation
- Chapterization
- Audio to LLM integration
Async API with FastAPI
Easy-to-use CLI interface
Configurable logging
CORS support
Health check endpoint

Quickstart with Claude Desktop

Get your API key from Gladia. There is a free tier available.
Install uv (Python package manager), install with curl -LsSf https://astral.sh/uv/install.sh | sh or see the uv repo for additional install methods.
Go to Claude > Settings > Developer > Edit Config > claude_desktop_config.json to include the following:

{
  "mcpServers": {
    "Gladia": {
      "command": "uvx",
      "args": ["gladia-mcp"],
      "env": {
        "GLADIA_API_KEY": "<insert-your-api-key-here>"
      }
    }
  }
}

If you're using Windows, you will have to enable "Developer Mode" in Claude Desktop to use the MCP server. Click "Help" in the hamburger menu at the top left and select "Enable Developer Mode".

Other MCP clients

For other clients like Cursor and Windsurf, run:

pip install gladia-mcp
python -m gladia_mcp --api-key={{PUT_YOUR_API_KEY_HERE}} --print to get the configuration. Paste it into appropriate configuration directory specified by your MCP client.

Example usage

Try asking Claude:

"Transcribe this audio file and identify different speakers"
"Convert this recording to text and translate it to Spanish"
"Analyze the sentiment and emotions in this speech"
"Extract key topics and create chapters from this long audio file"
"Transcribe this conversation and summarize the main points"

Optional features

You can add the GLADIA_MCP_BASE_PATH environment variable to the claude_desktop_config.json to specify the base path MCP server should look for and output files specified with relative paths.

Contributing

If you want to contribute or run from source:

Clone the repository:

git clone https://github.com/gladia/gladia-mcp
cd gladia-mcp

Create a virtual environment and install dependencies using uv:

uv venv
source .venv/bin/activate
uv pip install -e ".[dev]"

Copy .env.example to .env and add your Gladia API key:

cp .env.example .env
# Edit .env and add your API key

Run the tests to make sure everything is working:

./scripts/test.sh
# Or with options
./scripts/test.sh --verbose --fail-fast

Install the server in Claude Desktop: mcp install gladia_mcp/server.py
Debug and test locally with MCP Inspector: mcp dev gladia_mcp/server.py

API Endpoints

Health Check

GET /health

Transcribe Audio

POST /transcribe

Parameters:

file: Audio file (multipart/form-data)
diarization: Enable speaker diarization (boolean, optional)
language: Language code (string, optional)

Example using curl:

curl -X POST "http://localhost:8000/transcribe" \
  -H "accept: application/json" \
  -H "Content-Type: multipart/form-data" \
  -F "file=@audio.wav" \
  -F "diarization=true"

Troubleshooting

Logs when running with Claude Desktop can be found at:

Windows: %APPDATA%\Claude\logs\mcp-server-gladia.log
macOS: ~/Library/Logs/Claude/mcp-server-gladia.log

MCP Gladia: spawn uvx ENOENT

If you encounter the error "MCP Gladia: spawn uvx ENOENT", confirm its absolute path by running this command in your terminal:

which uvx

Once you obtain the absolute path (e.g., /usr/local/bin/uvx), update your configuration to use that path (e.g., "command": "/usr/local/bin/uvx"). This ensures that the correct executable is referenced.

Development

Running Tests

pytest

Code Style

The project follows PEP 8 style guide. Use flake8 for linting:

flake8 gladia_mcp

License

MIT License