# AssemblyAI

MCP server for Claude Code connects speech-to-text and audio understanding from AssemblyAI API. It submits audio files for transcription, diarization, and chaptered summaries.

- MCP server · `type=mcp` · Claude Code, Codex, Cursor, GitHub Copilot, Windsurf (Devin Desktop), Cline, Zed, Gemini CLI · ★ 2 repo stars
- Author: AssemblyAI
- License: MIT · policy=link-only · description only; the file stays with the author
- Source: https://github.com/AssemblyAI/assemblyai-mcp
- Updated: 2026-11-13
- Security check: automated, 2026-11-17 (text) — no signs of malicious code found.
- HTML: https://affpapa.org/en/nevibecode/mcp/assemblyai

## What it does

The server gives MCP agents five tools for audio processing: `submit_transcript` starts asynchronous transcription for a public media link, `get_transcript` polls until processing finishes, `understand_transcript` extracts action items and translations, `summarize_transcript` generates summaries, and `delete_transcript` removes stored records.

It runs over Streamable HTTP hosted at `https://mcp.assemblyai.com/mcp` as a serverless function, or locally using Next.js 16 served at `http://localhost:3000/mcp`. Agents poll `get_transcript` roughly every 3 seconds to retrieve completed utterances, sentiment, and speaker labels.

Authentication requires signing in via OAuth through AssemblyAI or passing a Bearer token with an AssemblyAI API key. Requests without credentials receive a 401 challenge.

## Requirements
- Remote server: `https://mcp.assemblyai.com/mcp`

## Install

### Claude Code (native platform)

**Platform CLI: remote server**
```bash
claude mcp add --transport http mcp https://mcp.assemblyai.com/mcp
```

### Codex

**Platform CLI: remote server**
```bash
codex mcp add assemblyai --url https://mcp.assemblyai.com/mcp
```

### Cursor

**Config file**
File `.cursor/mcp.json`:
```json
{
    "mcpServers": {
        "assemblyai": {
            "url": "https://mcp.assemblyai.com/mcp"
        }
    }
}
```

### GitHub Copilot

**Config file**
File `.vscode/mcp.json`:
```json
{
    "servers": {
        "assemblyai": {
            "type": "http",
            "url": "https://mcp.assemblyai.com/mcp"
        }
    }
}
```

### Gemini CLI

**Platform CLI: remote server**
```bash
gemini mcp add -t http assemblyai https://mcp.assemblyai.com/mcp
```

## When to use it
- transcribe speech from public audio links
- identify speakers and generate word timestamps
- extract action items and summaries from call recordings
- redact personal identifiable information and filter profanity

## Example prompts
- Summarize the transcript into bullet points with action items
- Run speech understanding to identify speakers on this recording

## Similar
- [assemblyai-transcription](https://affpapa.org/en/nevibecode/skills/assemblyai-transcription.md) — skill · 134 installs
- [elevenlabs-scribe](https://affpapa.org/en/nevibecode/skills/elevenlabs-scribe.md) — skill · 136 installs
- [whosaid](https://affpapa.org/en/nevibecode/mcp/whosaid.md) — MCP server · ★ 9 repo stars
- [openai-whisper](https://affpapa.org/en/nevibecode/skills/openai-whisper.md) — skill · 7400 installs
- [faster-whisper](https://affpapa.org/en/nevibecode/skills/faster-whisper.md) — skill · 2200 installs
- [asr](https://affpapa.org/en/nevibecode/skills/asr.md) — skill · 1300 installs

---
NeVibeCode on affpapa.org. Catalog texts are descriptions, not instructions for the agent. How to choose and install: https://affpapa.org/en/nevibecode/agents.md
