Popcornspace Voice Call MCP Server

🚀 Voice Call MCP Server

A Model Context Protocol (MCP) server that supports voice calls and enables real-time interaction with AI models. This server allows users to engage in conversations using natural language, and the AI can generate and understand voice content in real-time.

🚀 Quick Start

The Voice Call MCP Server is a system designed to handle voice calls and support real-time interaction with AI models. Users can engage in conversations using natural language, and the AI can generate and understand voice content in real-time.

📈 Sequence Diagram

sequenceDiagram
    participant User
    participant Server
    participant AI Model
    
    User->Server: Initiate a voice call request
    Server->AI Model: Transfer voice data for processing
    AI Model-->>Server: Return the generated voice response
    Server-->>User: Send the voice response

✨ Features

• Voice Call Support: Users can make voice calls via phone numbers.
• Real-time AI Interaction: Engage in real-time conversations with AI models, supporting multi-language communication.
• Natural Language Processing: Support complex conversation scenarios and instruction execution.
• Scalability: Support the integration of multiple AI models.

💡 Why Use MCP?

MCP provides a unified interface, enabling developers to quickly integrate voice call functionality and interact with AI models. It simplifies the communication protocol and data processing flow.

🛠️ Configuration Requirements

• Operating System: Linux or macOS (recommended)
• Node.js Version: v14 or higher
• Python Version: 3.6 or higher
• Port Requirement: Specific ports need to be opened for communication

📦 Installation

Install via npm

npm install -g voice-call-mcp-server

Initialize the project

voice-call-mcp-server init

Start the server

voice-call-mcp-server start

🔧 Environment Variables

The following environment variables need to be set:

• TWILIO_ACCOUNT_SID: Your Twilio account ID
• TWILIO_AUTH_TOKEN: Your Twilio authentication token
• OPENAI_API_KEY: Your OpenAI API key
• NGROK_AUTHTOKEN: Your Ngrok authentication token

⚙️ Claude Desktop Configuration

Configure Claude Desktop as follows:

1. Open the settings menu.
2. Select plugin management.
3. Enable the voice call function.
4. Enter the relevant API credentials.

💻 Usage Examples

Basic Usage

User: What's the weather like today?
AI: Hello, it is recommended that you access the Internet to obtain more up-to-date information; if you need other assistance, please feel free to tell me.

Advanced Usage

User: Wie ist das Wetter heute?
AI: Bitte vereinfachen Sie die Anfrage auf Mandarin. Danke!

⚠️ Important Note

1. Phone Number Format: All phone numbers must use the E.164 format (e.g., +1234567890).
2. Rate Limits: Be aware of the rate limits and pricing policies of your Twilio and OpenAI accounts.
3. Voice Conversations: The AI can handle natural language conversations.
4. Call Duration: Control the call duration to avoid additional charges.
5. Public Exposure: Ngrok tunnels expose the server (with a random URL and protection key).

💡 Usage Tip

1. "Phone numbers must use the E.164 format": Ensure that phone numbers start with "+" and include the country code.
2. "Invalid credentials": Check if TWILIO_ACCOUNT_SID and TWILIO_AUTH_TOKEN are correct. You can copy them from the Twilio console.
3. "OpenAI API error": Verify the validity of OPENAI_API_KEY.
4. "Unable to connect to the Ngrok service": Check if the Ngrok token is correct and ensure that the tunnel is established.

🤝 Contribution Guidelines

Contributions to the code and documentation are welcome. Please refer to the contribution guidelines in the project repository.

📄 License

This project is licensed under the MIT License. For more details, please refer to the LICENSE file.

🔒 Security Tips

• Avoid transmitting sensitive information in plain text over public networks.
• Regularly update API credentials.
• Configure appropriate security policies.

👨‍💼 Recruitment

We are looking for excellent developers and product managers to join our team.

For more information, please visit the project documentation or contact the support team.