MCP Audio

Transcribes audio via the Audio API, exposing an MCP tool for identify_voice with secure key management and help endpoints.

python

8

GitHub Stars

python

Language

6 months ago

First Indexed

2 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "aio-2030-mcp-audio": {
      "command": "python",
      "args": [
        "src/mcp_server.py"
      ],
      "env": {
        "API_KEY": "YOUR_API_KEY",
        "AUDIO_URL": "https://AUDIO_URL_PLACEHOLDER"
      }
    }
  }
}

You run an MCP server that offers voice-to-text transcription via the Audio speech recognition API. It exposes a transcription capability you can call from an MCP client using a standard tools interface, making it easy to convert audio inputs into text within your MCP workflows.

How to use

You integrate with the server from an MCP client by calling the available transcription tool through the standard tools interface. Your client will provide audio data, and the server will return the resulting transcript along with a confidence score. Use cases include turning user speech into searchable text, captioning audio content, and enabling voice-driven automation within your MCP workflows.

How to install

Prerequisites you need before installation are Python (with venv support) and Git. Optionally, you may use Docker for containerized runs.

Step 1: Clone the project and set up a Python virtual environment.

git clone git@github.com:AIO-2030/mcp-audio.git
cd mcp-audio
python -m venv venv && source venv/bin/activate
pip install -r requirements.txt

Step 2: Prepare environment variables. Copy the example and set your values for the audio API URL and API key.

cp .env.example .env

Then set: AUDIO_URL=https--xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx API_KEY=sk-xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx


Step 3: Run the MCP server locally.

python src/mcp_server.py


Step 4: Alternative Docker-based run. Build and run the container with the environment file.

docker build -t mcp-audio . docker run --env-file .env -p 8080:8080 mcp-audio

## Additional notes

The server includes endpoints for audio transcription and MCP tool exposure. It is designed to work with the AIO tool protocol and provides a help endpoint for discovery. Make sure you provide a valid AUDIO\_URL and API\_KEY in your environment for proper operation.

## Configuration and tools

Environment variables you control directly from your deployment include the audio API URL and the API key.

If you deploy via Docker, you can pass your environment variables through an env file to keep credentials out of the image.

## Testing and health

Optionally, run health checks or test scripts to ensure the audio transcription path works as expected. Use the provided test runners to verify base64 audio handling and overall MCP readiness.

## Available tools

### identify\_voice

Transcribes uploaded audio into text, returning the transcript and confidence score for the given audio input.

### tools.call

Supports the AIO protocol for invoking MCP tools with structured JSON-RPC inputs.

### help

Provides an accessible endpoint to discover MCP capabilities and usage details.