OpenAI Speech-to-Text transcriptions MCP Server
@Ichigo3766
A MCP server that provides audio transcription capabilities using OpenAI's Whisper API
Overview
What is OpenAI Speech-to-Text transcriptions MCP Server?
A Model Context Protocol server that provides audio transcription capabilities using OpenAI’s API. It is built for developers who want to integrate speech-to-text into MCP-compatible applications.
How to use OpenAI Speech-to-Text transcriptions MCP Server?
Clone the repository, run npm install and npm run build, then set your OPENAI_API_KEY environment variable. Add the server configuration to your MCP settings with the compiled index.js file. Use the transcribe_audio tool with a file path; optionally set save_to_file (Boolean) and language (ISO‑639‑1 code).
Key features of OpenAI Speech-to-Text transcriptions MCP Server
- Transcribes audio files via OpenAI’s API
- Optional language parameter (ISO‑639‑1 code)
- Option to save transcription output to a file
- Configurable OpenAI base URL and model
- Open source under the MIT License
Use cases of OpenAI Speech-to-Text transcriptions MCP Server
- Transcribing meeting recordings for searchable notes
- Adding speech-to-text to AI assistants or chatbots
- Processing user‑uploaded audio in MCP‑based apps
- Converting voice memos to text with language selection
FAQ from OpenAI Speech-to-Text transcriptions MCP Server
What does the server do?
It provides a transcribe_audio tool that sends audio files to OpenAI’s speech‑to‑text endpoint and returns the transcription.
What are the requirements?
You need Node.js, npm, and an OpenAI API key. Optionally you can set a custom base URL and model.
How do I configure the server?
Add the server to your MCP configuration with the command: "node" and args pointing to the built index.js, and set OPENAI_API_KEY in the env object.
What parameters does transcribe_audio accept?
It requires filepath (string). Optional parameters: save_to_file (Boolean) and language (ISO‑639‑1 code, e.g., “en”).
Where is my audio data processed?
The server sends audio to OpenAI’s API. No local storage or processing is described in the README.
