CleanAudio

Audio MCP server for AI agents

Expose CleanAudio cleaning and editing tools to Claude, Cursor, ChatGPT, and Claude Code through Model Context Protocol

MCP
1

Copy the CleanAudio connector URL

You'll paste this URL into Claude in the next step

2

Go to Claude → Customize

In Claude desktop or claude.ai, go to Customize → Connectors. Name it CleanAudio and paste the URL

Open Claude Customize
3

Connect, sign in and start

Sign in, then ask Claude to clean, cut, or transcribe audio

Start cleaning

An audio MCP server lets AI agents call audio tools over the Model Context Protocol. CleanAudio's remote MCP at https://mcp.cleanaudio.app/mcp exposes noise cleaning, cutting, joining, TTS, silence removal, and transcription after you sign in with a CleanAudio account.

Choose this page when you are comparing MCP audio backends or searching for a ready remote server instead of building your own.

What this audio MCP server includes

Professional microphone

Production audio pipeline

Not a toy stub. Tools map to CleanAudio's online processing used by podcasters and creators on the website.

Remote MCP architecture

Remote-first for agents

Streamable HTTP MCP means agents connect over the network. You do not ship a local audio binary with every install.

Multiple AI agents

Agent-agnostic connector

Same server powers Claude, Cursor, ChatGPT, Claude Code, and other MCP clients that support remote HTTP.

Use the CleanAudio audio MCP server

One URL works across MCP-compatible agents. Pick your client, paste the connector, authorize, then prompt.

  1. 1

    Copy the remote MCP URL

    https://mcp.cleanaudio.app/mcp

  2. 2

    Add it in your AI client

    Claude Connectors, Cursor MCP settings, ChatGPT connectors, or a Claude Code setup prompt all accept the same URL.

  3. 3

    Authorize CleanAudio

    Complete browser sign-in. Tool calls run as your user against the CleanAudio backend.

  4. 4

    Call audio tools from prompts

    Ask the agent to clean noise, edit timelines, generate speech, or transcribe. MCP handles discovery and invocation.

Connector URL and sample configs

Start with the URL, then use agent-specific samples as needed.

Remote MCP URLURL
https://mcp.cleanaudio.app/mcp
Cursor exampleJSON
{
  "mcpServers": {
    "cleanaudio": {
      "url": "https://mcp.cleanaudio.app/mcp"
    }
  }
}
Claude ConnectorsSteps
1. Claude → Settings → Connectors
2. Add custom connector
3. Paste: https://mcp.cleanaudio.app/mcp
4. Connect → sign in to CleanAudio
5. Done (no API key)

Tools your agent can call

Your agent can call clean_noise, cut_audio, join_audio, text_to_speech, remove_silence, and transcribe on the same CleanAudio backend as the web app.

clean_noise

Remove background noise from podcast recordings

cut_audio

Trim MP3 files to create ringtones or clips

join_audio

Merge multiple audio files into one

text_to_speech

Convert blog posts to audio narration

remove_silence

Clean up long recordings by removing pauses

transcribe

Convert speech to text automatically

CleanAudio Audio MCP Server FAQ

It is an MCP endpoint that exposes audio capabilities (clean, edit, speak, transcribe) so AI agents can call them like tools instead of inventing file processing themselves.

https://mcp.cleanaudio.app/mcp

Any client that supports remote HTTP MCP, including Claude Connectors, Cursor, ChatGPT connectors where available, and Claude Code.

No for the standard path. CleanAudio hosts the remote server. You connect and sign in.

The website is a visual workspace. The MCP server is for agents: same backend tools, invoked from chat or CLI agents after OAuth.

No. MCP is included with CleanAudio plans. Usage consumes your normal processing credits.

Connect your agent to CleanAudio MCP

Copy the remote URL, add it in your AI client, and process audio from prompts.

Open connector setupMCP overview