gettranscribe-mcp-server

get-transcribe/gettranscribe-mcp-server

3.2

If you are the rightful owner of gettranscribe-mcp-server and would like to certify it and/or have it hosted online, please leave a comment on the right or send an email to henry@mcphub.com.

MCP server for GetTranscribe, an AI-powered video transcription service, enabling interaction with transcriptions through AI assistants.

Tools
6
Resources
0
Prompts
0

GetTranscribe MCP Server

GetTranscribe is an AI-powered video transcription service that delivers accurate text transcriptions from Instagram, TikTok, YouTube, Twitter, Pinterest, and Google Drive videos in seconds. Trusted by 12,000+ creators and businesses worldwide, GetTranscribe offers 95%+ accuracy, bulk processing, folder organization, API access, and integrations with n8n, Make, Chrome extension, and mobile apps for iOS and Android.

MCP (Model Context Protocol) server for GetTranscribe. This allows you to interact with your GetTranscribe transcriptions through AI assistants like Claude, ChatGPT, and other MCP-compatible clients.

Supports both MCP transports:

  • stdio - For standard MCP client integration (Claude Desktop, VS Code, etc.)
  • Streamable HTTP - For web-based and advanced clients (ChatGPT, remote access)

📖 See for detailed transport comparison and usage guide.

🔗 Client-Specific Setup

🤖 ChatGPT Users

ChatGPT requires OAuth 2.0 authentication. See for complete setup guide.

🔧 OpenAI Playground Users

Step-by-Step Setup:

  1. Open OpenAI Playground

  2. Add MCP Server

    • Click on "Chat" in the left sidebar
    • Click "+ Add" under "Tools" section
    • Select "MCP Server"
    • Click on " + Server"

  3. Configure GetTranscribe Server

    • Name: GetTranscribe
    • URL: https://gettranscribe-mcp-server.onrender.com/mcp
    • API Key / Auth token: Select "API Key" from dropdown
    • Enter your GetTranscribe API key: your_api_key_here

  4. Select Tools

    • Enable all available tools:
      • ✅ create_transcription
      • ✅ get_transcription
      • ✅ list_transcriptions
      • ✅ create_transcription_folder
      • ✅ get_transcription_folder
      • ✅ list_transcription_folders

  5. Save and Test

    • Click "Add" to save the configuration
    • Start chatting and ask: "List my recent transcriptions"

🎯 Cursor IDE & Other MCP Clients

🌟 Recommended: SSE Transport (HTTP) Add to your ~/.cursor/mcp.json:

{
  "mcpServers": {
    "gettranscribe": {
      "url": "https://gettranscribe-mcp-server.onrender.com/mcp",
      "headers": {
        "Authorization": "Bearer your_gtr_api_key"
      }
    }
  }
}

Alternative: stdio Transport

{
  "mcpServers": {
    "gettranscribe": {
      "command": "npx",
      "args": ["-y", "gettranscribe-mcp@latest"],
      "env": {
        "GETTRANSCRIBE_API_KEY": "your_gtr_api_key"
      }
    }
  }
}

For Development (Local Server):

{
  "mcpServers": {
    "gettranscribe-local": {
      "url": "http://localhost:8080/mcp",
      "headers": {
        "Authorization": "Bearer your_gtr_api_key"
      }
    }
  }
}

🚀 Quick Start

Installation

npm install -g gettranscribe-mcp-server

Configuration

For stdio transport (most common):

Add to your MCP client configuration (e.g., ~/.cursor/mcp.json):

{
  "mcpServers": {
    "gettranscribe": {
      "command": "gettranscribe-mcp",
      "env": {
        "GETTRANSCRIBE_API_KEY": "your_api_key_here"
      }
    }
  }
}
For Streamable HTTP transport:

Start the server:

MCP_TRANSPORT=http PORT=8080 GETTRANSCRIBE_API_KEY=your_key gettranscribe-mcp

Or use npm scripts:

npm run start:http        # General HTTP mode
npm run start:chatgpt     # Alias for ChatGPT users
npm run dev               # Development mode

Then configure your client to connect to http://localhost:8080/mcp

🤖 ChatGPT Integration

This server is fully compatible with ChatGPT connectors and deep research. It implements the required search and fetch tools.

📖 See for detailed ChatGPT setup instructions.

Get Your API Key

  1. Visit GetTranscribe.ai
  2. Create an account or sign in
  3. Go to Settings → API Keys
  4. Generate a new API key
  5. Copy it to your MCP configuration

🛠️ Available Tools

Transcription Tools

  • create_transcription - Create transcription from video URL (YouTube, Instagram, TikTok, Meta)
  • get_transcription - Retrieve specific transcription by ID
  • list_transcriptions - List your transcriptions with filtering

Folder Management Tools

  • create_transcription_folder - Create folders to organize transcriptions
  • get_transcription_folder - Get folder details and contents
  • list_transcription_folders - List your folders

🎯 Cursor IDE Setup Guide

Step 1: Open Cursor Configuration

Open ~/.cursor/mcp.json in your editor (create if it doesn't exist).

Step 2: Add GetTranscribe Configuration

🌟 Recommended (SSE/HTTP):

{
  "mcpServers": {
    "gettranscribe": {
      "url": "https://api.gettranscribe.ai/mcp",
      "headers": {
        "Authorization": "Bearer gtr_your_api_key_here"
      }
    }
  }
}

Step 3: Restart Cursor

Restart Cursor IDE to load the new MCP server.

Step 4: Test the Connection

In Cursor chat, try: "List my recent transcriptions"

💬 Example Usage

After setup, you can ask your AI assistant:

⏱️ Transcription Segments

When creating transcriptions, you can request timestamped segments by setting include_segments: true. This provides the transcription broken down by time intervals:

Basic transcription:

✅ Transcription created successfully!
**ID:** 492
**Transcription:**
Welcome to our video about AI and machine learning...

With segments:

✅ Transcription created successfully!
**ID:** 492
**Transcription:**
Welcome to our video about AI and machine learning...

**Transcription Segments:**
[0:00 - 0:05] Welcome to our video about AI
[0:05 - 0:12] and machine learning. Today we'll cover
[0:12 - 0:18] the fundamental concepts you need to know.

🔧 Tool Parameters

create_transcription

{
  "url": "https://www.instagram.com/reel/example", // Required: Video URL
  "folder_id": 123,                                // Optional: Folder ID
  "prompt": "Focus on key takeaways",              // Optional: Custom prompt
  "language": "en",                                // Optional: Language code
  "include_segments": true                         // Optional: Include timestamped segments
}

get_transcription

{
  "transcription_id": 123                          // Required: ID of the transcription to retrieve
}

list_transcriptions

{
  "folder_id": 123,                               // Optional: Filter by folder
  "platform": "youtube",                         // Optional: Filter by platform (instagram, tiktok, youtube, meta)
  "limit": 10,                                   // Optional: Results limit (default: 10, max: 50)
  "skip": 0                                      // Optional: Pagination offset
}

create_transcription_folder

{
  "name": "My Folder",                           // Required: Folder name
  "parent_id": 456                               // Optional: Parent folder ID for nested structure
}

get_transcription_folder

{
  "folder_id": 123                               // Required: ID of the folder to retrieve
}

list_transcription_folders

{
  "parent_id": 456,                              // Optional: Filter by parent folder ID (null for root folders)
  "limit": 10,                                   // Optional: Results limit (default: 10)
  "skip": 0                                      // Optional: Pagination offset
}

🌐 Environment Variables

  • GETTRANSCRIBE_API_KEY (required) - Your GetTranscribe API key
  • GETTRANSCRIBE_API_URL (optional) - API endpoint (defaults to https://gettranscribe.ai)
  • MCP_TRANSPORT (optional) - Transport mode: "stdio" (default) or "http"
  • PORT (optional) - HTTP server port (default: 8080, only for HTTP transport)
  • MCP_PATH (optional) - HTTP endpoint path (default: "/mcp", only for HTTP transport)

🔧 Transport Configuration

📊 Quick Reference

TransportUse ForConfigurationURL
SSE (HTTP) ⭐Cursor, MCP ClientsURL + headershttps://api.gettranscribe.ai/mcp
stdioClaude Desktop, VS CodeCommand + envN/A
HTTP LocalDevelopment, ChatGPTnpm run start:httphttp://localhost:8080/mcp

⭐ SSE is recommended for most applications as it provides better performance and real-time capabilities.

📟 stdio Transport (Default - for MCP Clients)

For Claude Desktop, VS Code, and other MCP clients:

# Default mode - starts in stdio
node mcp-server.js

# Or explicitly specify stdio
MCP_TRANSPORT=stdio node mcp-server.js

# With API key
GETTRANSCRIBE_API_KEY=your_key node mcp-server.js

Use Case: Standard MCP client integration

🌐 HTTP Transport (for ChatGPT, OpenAI API, Web Clients)

For ChatGPT connectors, OpenAI API, and web-based access:

# Start HTTP server on port 8080
MCP_TRANSPORT=http PORT=8080 node mcp-server.js

# With API key (optional, can use headers)
MCP_TRANSPORT=http PORT=8080 GETTRANSCRIBE_API_KEY=your_key node mcp-server.js

# Custom port and endpoint
MCP_TRANSPORT=http PORT=3000 MCP_PATH=/api/mcp node mcp-server.js

# Production mode
MCP_TRANSPORT=http PORT=8080 GETTRANSCRIBE_API_URL=https://api.gettranscribe.ai node mcp-server.js

Use Case: ChatGPT connectors, OpenAI API, browser-based clients

🚀 Quick Start Scripts

# For MCP clients (Claude, VS Code)
npm run start:stdio

# For ChatGPT and OpenAI API
npm run start:http

# For development
npm run dev

📡 HTTP Transport Features

The HTTP transport includes:

  • SSE (Server-Sent Events) for real-time communication
  • OAuth 2.0 support for ChatGPT integration
  • Multiple auth methods: x-api-key header, Bearer token, OAuth JWT
  • Session management with Mcp-Session-Id headers
  • Health check endpoint at /health
  • CORS support for browser clients

Note: HTTP transport automatically includes SSE support. When you start with MCP_TRANSPORT=http, the server handles both regular HTTP requests and SSE streams on the same endpoint.

📋 Supported Platforms

  • YouTube
  • Instagram (posts, reels, stories)
  • TikTok
  • Meta/Facebook

🔐 Authentication & Privacy

Authentication Methods

📱 ChatGPT Users:

  • OAuth 2.0 flow (required by ChatGPT)
  • Users authorize with their GetTranscribe API key through a secure web form
  • Access tokens are JWT-based with 24-hour expiration

🔧 OpenAI API Users:

  • x-api-key header authentication OR
  • Authorization: Bearer <api_key> authentication
  • No OAuth flow needed for programmatic access

🎯 MCP Clients (Claude, etc.):

  • Environment variable authentication
  • API key set in client configuration

Privacy & Security

  • Each user connects with their own API key
  • All data remains private to your GetTranscribe account
  • Secure API communication with HTTPS
  • No data is stored by the MCP server
  • OAuth tokens are temporary and automatically expire

📝 License

MIT License - see LICENSE file for details

🤝 Support

🔄 Version History

1.1.0

  • ✨ NEW: Added include_segments parameter to create_transcription
  • ✨ NEW: Timestamped segment support with formatted output
  • 📝 Enhanced response format with full transcription text
  • 🛠️ Improved error handling for segment parsing

1.0.0

  • Initial release
  • Support for all 6 core tools
  • Full MCP protocol compatibility
  • Authentication via API keys