Notta MCP Server

Model Context Protocol (MCP) server for audio/video transcription and record management through Notta.

What is MCP?

Model Context Protocol is a standard for connecting AI assistants (like Claude Desktop, Cursor, etc.) to external tools and data sources.

Installation

npm install -g @notta-labs/notta-mcp

Configuration

Add to your MCP client configuration (e.g., Claude Desktop's claude_desktop_config.json):

{
  "mcpServers": {
    "notta": {
      "command": "notta-mcp"
    }
  }
}

Configuration File Locations

| Client | Config Path | |--------|-------------| | Claude Desktop (macOS) | ~/Library/Application Support/Claude/claude_desktop_config.json | | Claude Desktop (Windows) | %APPDATA%\Claude\claude_desktop_config.json | | Cursor | See Cursor MCP documentation |

Authentication

Before using the MCP server, authenticate via the CLI:

# Install CLI tool (if not already installed)
npm install -g @notta-labs/notta-cli

# Authenticate (opens browser for OAuth)
notta auth setup

The MCP server will use the same credentials stored at ~/.config/notta_cli/credentials.json.

Available Tools

The MCP server exposes the following tools to AI assistants:

| Tool | Description | Parameters | |------|-------------|------------| | notta_transcribe | Upload and transcribe audio/video file | file_path, language (optional) | | notta_status | Check transcription status | record_id, wait (optional) | | notta_view | Get transcription result | record_id | | notta_search | Search records by keyword | keyword, page, page_size | | notta_list | List records and folders | group_id, page, page_size, order_by, order_type |

Supported Languages

zh-CN zh-TW zh-HK en-US ja-JP ko-KR vi-VN th-TH hi-IN id-ID jv-ID ms-MY fil-PH ne-NP pt-PT es-ES fr-FR de-DE it-IT nl-NL ru-RU uk-UA pl-PL cs-CZ sk-SK sl-SI hr-HR sr-RS bg-BG ro-RO hu-HU lt-LT lv-LV da-DK nb-NO sv-SE fi-FI is-IS tr-TR ar-AE ar-EG fa-IR he-IL hy-AM az-AZ ka-GE el-GR gu-IN kn-IN ta-IN te-IN ur-IN bn-IN km-KH lo-LA si-LK sw-KE zu-ZA af-ZA am-ET eu-ES ca-ES gl-ES su-ID

Default language is ja-JP.

Example Usage

Once configured, you can ask your AI assistant:

• "Transcribe this meeting recording: /path/to/meeting.mp4"
• "Check the status of record 12345"
• "Show me the transcription for record 12345"
• "Search my recordings for 'quarterly review'"
• "List all my recordings"