Voice Recorder (Whisper) MCP server

This MCP server allows you to record audio and transcribe it using OpenAI's Whisper model. It works either as a standalone server or as a custom extension for the Goose AI agent, giving you flexible options for voice-to-text capabilities.

Installation

You can install the Voice Recorder MCP server directly from source:

git clone https://github.com/DefiBax/voice-recorder-mcp.git
cd voice-recorder-mcp
pip install -e .

Running the Server

Basic Usage

Start the server with default settings (using the base.en Whisper model):

voice-recorder-mcp

Advanced Options

Specify a different Whisper model for better accuracy:

voice-recorder-mcp --model medium.en

Adjust the audio sample rate if needed:

voice-recorder-mcp --sample-rate 44100

Testing with MCP Inspector

For interactive testing of your server:

# Install the MCP Inspector
npm install -g @modelcontextprotocol/inspector

# Run your server with the inspector
npx @modelcontextprotocol/inspector voice-recorder-mcp

Integration with Goose AI Agent

1. Open Goose and navigate to Settings > Extensions > Add > Command Line Extension

2. Set the name to voice-recorder

3. In the Command field, enter the full path to the executable:

   /full/path/to/voice-recorder-mcp
   

   For a specific model:

   /full/path/to/voice-recorder-mcp --model medium.en
   

   To find the correct path, run:

   which voice-recorder-mcp
   

4. Start a conversation with Goose and introduce the recorder with: "I want you to take action from transcriptions returned by voice-recorder. For example, if I dictate a calculation like 1+1, please return the result."

Available Tools

The server provides three main functions:

• start_recording: Begins recording audio from your default microphone
• stop_and_transcribe: Stops the current recording and converts it to text
• record_and_transcribe: Records for a specified duration and automatically transcribes

Whisper Model Options

You can choose from several Whisper models depending on your needs:

Models with the .en suffix are optimized for English content.

Configuration via Environment Variables

You can also configure the server using environment variables:

# Set Whisper model
export WHISPER_MODEL=small.en

# Set audio sample rate
export SAMPLE_RATE=44100

# Set maximum recording duration (seconds)
export MAX_DURATION=120

# Then run the server
voice-recorder-mcp

Troubleshooting

Common Issues

• No audio being recorded: Verify your microphone permissions and system settings
• Model download errors: Ensure you have a stable internet connection for the initial model download
• Integration with Goose: Double-check that the command path is correct
• Audio quality issues: Try adjusting the sample rate (default: 16000)

How to add this MCP server to Cursor

There are two ways to add an MCP server to Cursor. The most common way is to add the server globally in the ~/.cursor/mcp.json file so that it is available in all of your projects.

If you only need the server in a single project, you can add it to the project instead by creating or adding it to the .cursor/mcp.json file.

Adding an MCP server to Cursor globally

To add a global MCP server go to Cursor Settings > MCP and click "Add new global MCP server".

When you click that button the ~/.cursor/mcp.json file will be opened and you can add your server like this:

{
    "mcpServers": {
        "cursor-rules-mcp": {
            "command": "npx",
            "args": [
                "-y",
                "cursor-rules-mcp"
            ]
        }
    }
}

Adding an MCP server to a project

To add an MCP server to a project you can create a new .cursor/mcp.json file or add it to the existing one. This will look exactly the same as the global MCP server example above.

How to use the MCP server

Once the server is installed, you might need to head back to Settings > MCP and click the refresh button.

The Cursor agent will then be able to see the available tools the added MCP server has available and will call them when it needs to.

You can also explictly ask the agent to use the tool by mentioning the tool name and describing what the function does.