1
Copy the CleanAudio connector URL
You'll paste this URL into Claude in the next step
A hosted FFmpeg alternative for agent audio tasks
Connect to CleanAudio's hosted MCP, sign in with your account, and use predictable audio tools without installing, securing, or self-hosting FFmpeg wrappers
MCP
2
Go to Claude → Customize
In Claude desktop or claude.ai, go to Customize → Connectors. Name it CleanAudio and paste the URL
3
Connect, sign in and start
Sign in, then ask Claude to clean, cut, or transcribe audio
CleanAudio is a hosted alternative to building your own FFmpeg wrapper for common agent audio tasks. It offers named tools for clean_noise, cut_audio, join_audio, remove_silence, text_to_speech, and transcribe through a remote MCP connector.
It is not a promise of every FFmpeg codec or command. Use CleanAudio when you want hosted auth, usage billing, and predictable audio tools instead of operating a self-hosted media service.
Think of CleanAudio as Stripe for audio processing: hosted authentication, account-based usage billing, and predictable named tools instead of a self-hosted FFmpeg wrapper that each agent must operate.
What this MCP server does
- Clean background noise with clean_noise
- Cut an uploaded recording with cut_audio
- Join ordered clips with join_audio
- Remove pauses with remove_silence
- Generate speech with text_to_speech
- Create a transcript with transcribe
Example prompts
Ask Claude, Cursor, ChatGPT, or VS Code in plain language.
- “Remove background noise from this support call.”
- “Trim the first 30 seconds from this recording.”
- “Join these audio clips in order.”
- “Remove silence from this spoken lesson.”
- “Generate TTS narration for this paragraph.”
- “Transcribe this meeting recording.”
Why agents use hosted tools instead of FFmpeg wrappers
Named tools, clear outcomes
Call clean_noise, cut_audio, join_audio, remove_silence, text_to_speech, or transcribe instead of assembling shell commands.
Hosted auth and billing
Account sign-in and CleanAudio plan usage replace local secrets and custom billing plumbing.
No local media service
The remote connector avoids managing an FFmpeg installation or a long-running self-hosted wrapper.
Replace a local FFmpeg workflow with hosted MCP tools
Use the remote connector for supported audio actions, not as a claim to every FFmpeg command or codec.
- 1
Copy the hosted connector URL
Copy https://mcp.cleanaudio.app/mcp. CleanAudio hosts the MCP server, so there is no local server or FFmpeg installation.
- 2
Connect your AI agent
Add the URL in Claude Connectors, Cursor Settings → MCP, ChatGPT connectors, or your VS Code MCP client.
- 3
Sign in to CleanAudio
Complete the browser sign-in with your CleanAudio account. Authorization keeps processing and usage tied to your account.
- 4
Ask for a supported audio task
Upload or reference your audio, then give the agent a concrete instruction. It can choose the relevant CleanAudio tool.
Tools your agent can call
CleanAudio provides focused hosted tools for clean_noise, cut_audio, join_audio, text_to_speech, remove_silence, and transcribe. It does not claim universal FFmpeg command coverage.
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
Hosted MCP alternatives and setup guides
CleanAudio MCP hub
Connect an AI agent to CleanAudio's hosted audio tools.
MCP connector documentation
See setup guidance and the available CleanAudio MCP tools.
Claude MCP for audio
Use the same hosted connector from Claude.
Cursor MCP for audio
Run CleanAudio tools from the Cursor agent.
Hosted audio MCP
Explore another focused CleanAudio MCP workflow.
CleanAudio Hosted FFmpeg Alternative FAQ
No. It is a hosted alternative for specific supported audio actions, not every codec, filter, or shell command available in FFmpeg.
Use clean_noise, cut_audio, join_audio, remove_silence, text_to_speech, and transcribe through the MCP.
No. CleanAudio hosts the remote MCP endpoint and its processing backend.
Sign in with your CleanAudio account. Processing follows your CleanAudio plan and its usage model.
Do not assume arbitrary conversion support from the current MCP tool list. Use relevant CleanAudio web tools when conversion is needed.
Use a remote HTTP MCP client such as Claude, Cursor, ChatGPT connectors where available, or a compatible VS Code client.
Start hosted audio processing from your agent
Add the hosted CleanAudio MCP connector, sign in once, then send a clear audio request from your preferred agent.