Carbon Voice — Claude Code MCP server for voice messages
Carbon Voice handles voice messages with AI transcription, summaries, and action item extraction. Use it to support asynchronous communication.
no license updated 2026-10-29
Open on GitHubWhat Carbon Voice does
This MCP server processes voice messages. It produces AI transcriptions, concise summaries, and action items drawn from the recording. A voice note can become both readable text and a task list.
Carbon Voice is listed for Claude Code, Codex, Cursor, GitHub Copilot, Windsurf (Devin Desktop), Cline, Zed, and Gemini CLI. The Claude Code listing has no installation command, so you’ll need to find setup instructions for your platform before connecting it.
What it helps with
- transcribe a voice message
- summarize a recording
- extract action items from a voice note
- exchange messages asynchronously
Example prompts for the agent
- Transcribe this voice message and pull out the action items.
- Give me a short summary of this recording.
- Turn this voice note into text and a list of tasks.
How to install Carbon Voice
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 Carbon Voice
How do I install Carbon Voice in Claude Code?
Carbon Voice is listed for Claude Code, but there is no installation command here.
What does Carbon Voice do?
It processes voice messages to create AI transcriptions and summaries, then extracts action items.
Which platforms does Carbon Voice support?
The listed platforms are Claude Code, Codex, Cursor, GitHub Copilot, Windsurf (Devin Desktop), Cline, Zed, and Gemini CLI.