Voice Recorder MCP Server

A voice recording and transcription MCP server based on the Whisper model, which can be used as an extension of Goose AI or run independently, supporting various recording scenarios and model configurations.

Voice processing Education and learning tools #Voice Transcription #AI Extension #Recording Tool #Whisper Model .Python

rating : 2.5 points

downloads : 7.6K

update time : 2025-04-23

Open Site

What is Voice Recorder MCP?

Voice Recorder MCP is a multi - functional MCP server used to record audio and transcribe it through OpenAI's Whisper model. It can be used as an extension of the Goose AI agent or run independently.

How to use Voice Recorder MCP?

Simply install and configure it, and then you can start recording and transcribing audio. It supports various scenarios, such as daily conversations and meeting records.

Applicable Scenarios

Suitable for scenarios such as meeting records, study notes, daily voice tasks, and phone recordings.

Main Features

Record Audio

Record high - quality audio through the default microphone.

Transcribe Audio

Automatically transcribe the recorded audio into text using the Whisper model.

Goose AI Integration

As an extension tool for Goose AI, it supports real - time interaction and automated tasks.

Model Selection

Supports multiple Whisper models, allowing you to balance speed and accuracy according to your needs.

Advantages

Supports multiple Whisper models to meet different needs.

Easy to install and configure, suitable for all types of users.

Can be integrated into Goose AI for more efficient voice task processing.

Free and open - source, with rich community support.

Limitations

You may need to download large - scale model files when running it for the first time.

It has a certain dependence on the network environment (especially for the initial model download).

How to Use

Install the Server

Clone the project repository and install the dependencies.

Start the Server

Run the command to start the MCP server.

Configure the Goose AI Extension

Add the extension in the Goose AI settings and specify the server path.

Usage Examples

Meeting Records

Use Voice Recorder MCP to record the discussion content in a meeting and automatically generate text.

Study Notes

Record the course explanation and convert it into text for easy review.

Frequently Asked Questions

Why can't my microphone work properly?

How to switch between different Whisper models?

How to solve the problem of model download failure when running it for the first time?

Related Resources

Official GitHub Repository

Project source code and documentation.

Goose AI Official Website

Introduction to the Goose AI platform.

Whisper Model Documentation

Detailed description of the Whisper model.

🚀 Voice Recording MCP Server

A MCP server that uses OpenAI's Whisper model to record audio and transcribe it into text. Designed to run as a custom extension for Goose or as a standalone MCP server.

🚀 Quick Start

The Voice Recording MCP Server allows you to record audio from your microphone and transcribe it using OpenAI's Whisper model. It can be integrated with Goose AI or run independently.

✨ Features

Record audio from the default microphone.
Transcribe recordings using the Whisper model.
Integrate as a custom extension into the Goose AI agent.
Provide prompts for common recording scenarios.

📦 Installation

# Install from source code
git clone https://github.com/DefiBax/voice-recorder-mcp.git
cd voice-recorder-mcp
pip install -e .

💻 Usage Examples

Basic Usage

Run as a Standalone MCP Server

# Run with default settings (base.en model)
voice-recorder-mcp

# Run with a specified Whisper model
voice-recorder-mcp --model medium.en

# Adjust the sampling rate
voice-recorder-mcp --sample-rate 44100

Test with MCP Inspector

The MCP Inspector provides an interactive interface for testing the server:

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

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

Use with Goose AI Agent

Open Goose, go to Settings > Extensions > Add > Command-line Extension.
Set the name to voice-recorder.
Enter the full path to the Voice Recorder MCP executable in the command field:
```
/full/path/to/voice-recorder-mcp
```
Or run with a specified model:
```
/full/path/to/voice-recorder-mcp --model medium.en
```
To find the path:
```
which voice-recorder-mcp
```
No additional environment variable configuration is required for basic functionality.
In a conversation with Goose, introduce the recorder with the following statement: "I want you to take actions based on the transcribed text returned by the voice recorder. For example, when I dictate a calculation like 1+1, please return the result."

Advanced Usage

The following tools are available for use:

start_recording: Start recording audio from the default microphone.
stop_and_transcribe: Stop recording and transcribe the audio to text.
record_and_transcribe: Record for a specified duration (in seconds) and transcribe.

📚 Documentation

Available Whisper Models

This extension supports various Whisper model sizes:

Model	Speed	Accuracy	Memory Usage	Use Case
`tiny.en`	Fastest	Lowest	Smallest	Testing, quick transcription
`base.en`	Fast	Good	Low	Daily use (default)
`small.en`	Medium	Better	Medium	Balanced performance
`medium.en`	Slow	High	High	Important recordings
`large`	Slowest	Highest	High	Critical transcriptions

The .en suffix indicates that the model is optimized for English, making it faster and more accurate for English content.

Configuration Requirements

Python 3.12+
An audio input device (microphone)

Configuration Options

You can configure the server through environment variables:

# Set the Whisper model
export WHISPER_MODEL=small.en

# Set the audio sampling rate
export SAMPLE_RATE=44100

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

# Run the server with the following command:
voice-recorder-mcp

⚠️ Important Note

Ensure that microphone permissions are correctly set so that the application can access the microphone.
For network connectivity, ensure that the device has a stable internet connection to support model inference.
If you encounter any issues during use, check the console output or refer to the relevant documentation.

🤝 Contribution

Contributions from the community are welcome! Please refer to Contributing.md in the project repository to learn how to contribute to the development.