mcp-transcribe — Claude Code MCP server: what it does, how to install
Connects AI assistants to the Transcribe platform to convert audio and video into text. Handles both local files and remote links directly in chat.
Open on GitHubWhat mcp-transcribe does
The server integrates Transcribe.com features directly into your coding assistant. It lets the model read transcripts from media files and search previous texts. Old records can be renamed or deleted as needed. Transcribe supports over 100 languages. It handles background noise while generating word-level timestamps.
Four tools come out of the box. The convert-to-text utility processes recordings and delivers finished text immediately. Searching and retrieving existing transcripts is handled by read-transcriptions. You can modify or remove items using update-transcription, while get-balance monitors your remaining quota.
Local operation accepts local file paths alongside public URLs. You need node and git present on your system. Running convert-to-text consumes account time credits, so transcription fails when your balance runs out.
What it helps with
- convert local audio or video files into plain text
- extract speech transcripts with word timestamps and speaker labels
- search existing transcription records stored in your account
- track remaining audio processing credits from the editor
Example prompts for the agent
- Transcribe
meeting-recording.mp4and separate the speakers - Check my account minutes balance with
get-balance - Search my past transcripts for mentions of database migration
How to install mcp-transcribe
Install
The MCP registry has no launch command for it: see the repository README
Step by step, with commands for each platform: How to add an MCP server to Claude Code, Codex, Cursor, VS Code, Gemini CLI and more
Questions about mcp-transcribe
How do I install mcp-transcribe in Claude Code?
Verify that Node.js and Git are installed. Add the server to your assistant configuration using command npx with args ["-y", "github:transcribe-app/mcp-transcribe"] and set your integration URL in the MCP_INTEGRATION_URL environment variable.
What does mcp-transcribe do?
It provides tools for language models to transcribe audio recordings, inspect account balances, and search existing transcripts. The server handles both local file paths and online URLs.
Do I need local speech-to-text models or Python packages?
No local machine learning models or Python packages are required. Transcription runs through the Transcribe cloud service, requiring only Node.js to execute the MCP server.
