ForkFree

MCP servers

earshot

by Leonxlnx

Earshot is a toolkit that lets AI coding agents analyse and describe audio, check how well sound fits a description or a video, and create their own sound effects in code. It runs locally on a CPU as a command-line tool, a Python library or an MCP server.

New and promisingUnder two months old. Expect rough edges and fast changes.

View on GitHub

Stars
3.3k +92 a day
Licence
MIT
Runs on
Your computer
Setup
Moderate
Language
Python
Age
15 days

What you can do with it

  • Get a written description of an audio file and check how well it matches a phrase such as 'a deep mechanical keyboard press'
  • Review a launch video's soundtrack against a profile of professional films, with timestamped fixes
  • Render sound effects from a bank of 252 code-made sounds and place them to match cuts in a film

Make money with it

A freelancer could offer sound design and audio quality checks for product launch videos, using the tool to generate and test effects quickly.

An idea, not financial advice. Check the licence before charging for anything built on it.

Ways a business could use it

  • Check and improve the sound design of product demos and launch videos without hiring a sound designer
  • Generate short interface sounds and effects in code for apps, games or videos, avoiding stock sound library fees

Can you use it commercially?

Free to use, change and sell, including in paid products. Keep the copyright notice.

Not legal advice. Read the project's licence file before building a product on it.

Getting it running

It is a Python package used through a command-line tool, a Python API or an MCP server, and it needs to download models. The excerpt does not show the exact install steps.

Follow the official instructions in the project README.

Things to check

  • Each command-line call loads models and takes about 20 to 40 seconds, so the MCP server is better for many calls
  • The optional audio-LLM critique needs your own Gemini key and sends audio to an outside service
  • Fairly new project, so it may change quickly

Who's talking about it

Checked 29 Oct 2026. Summary drafted by AI from the project's README.

More in MCP servers

kanzo emanuele-em
Hidden gem

Kanzo is a small kanban board that runs in your browser and also works as an MCP server, so you and your AI coding agents share the same task list. All data sits in one local folder, which can be synced between machines through a private git repository.

138 stars +28 a day, est. AGPL-3.0
mcp-for-blender ahujasid
Worth a look

This is a Model Context Protocol server and Blender add-on that lets an AI assistant such as Claude control Blender 3D. It is for building and changing 3D scenes, objects and materials by typing prompts instead of working by hand.

32.5k stars MIT
token-optimizer-mcp ooples
Worth a look

Token Optimizer MCP is a Model Context Protocol server that compresses and caches the context sent to AI coding agents, and records how many tokens each agent saves. It is meant to lower the cost of using tools such as Claude Code and Gemini CLI.

563 stars MIT