whatisgithub

What is airfm?

dam0ntang/airfm — explained in plain English

Analysis updated 2026-05-18

22TypeScriptAudience · vibe coderComplexity · 4/5Setup · hard

In one sentence

A locally-run AI personal radio station that chats about music, recommends songs, and generates AI-voiced DJ commentary between tracks.

Mindmap

mindmap
  root((repo))
    What it does
      AI DJ radio
      Chat music requests
      AI voice commentary
    Tech stack
      TypeScript
      Claude
      Fish Audio
    Use cases
      Personalized music station
      Continuous FM broadcast
      Song recommendations
    Audience
      Music hobbyists
      Vibe coders
    Setup
      Claude CLI
      Fish Audio key
      NetEase API

Code map

Detail Auto

An interactive map of this repo's files and how they connect — its source is parsed live in your browser. Click Visualize to build it.

filefunction / class

What do people build with it?

USE CASE 1

Run a personalized AI radio station that recommends and plays music you like.

USE CASE 2

Listen to continuous FM-style broadcasts with AI-generated DJ commentary between songs.

USE CASE 3

Chat with the app to request specific songs, artists, or moods.

What is it built with?

TypeScriptClaudeFish AudioPWA

How does it compare?

dam0ntang/airfmaerdelan/housand-domaintoolmatrixalibaba/webmcp-nexus
Stars222222
LanguageTypeScriptTypeScriptTypeScript
Setup difficultyhardeasymoderate
Complexity4/52/53/5
Audiencevibe coderdeveloperdeveloper

Figures from each repo's GitHub metadata at analysis time.

How do you get it running?

Difficulty · hard Time to first run · 1h+

Needs a Claude CLI install, a Fish Audio key, and a self-hosted NetEase Music API instance.

No license information is stated in the README.

So what is it?

AirFM is a locally-run personal radio station application powered by AI. You run it on your own computer, and it acts as a personalized DJ: it can chat with you about what music you want to hear, recommend songs, and in FM mode, automatically generate and play AI-voiced commentary between tracks. The DJ voice is synthesized using a text-to-speech service called Fish Audio, and the AI behind the recommendations runs through Claude, an AI assistant made by Anthropic. The README is written in Chinese. Music playback is sourced from NetEase Cloud Music, a major Chinese music streaming service. You interact with the app through a web interface that can also be installed as a small desktop application via the Progressive Web App (PWA) standard, meaning it installs from your browser without going through an app store. The project has two main modes. In Chat mode, you type a music request and the app returns a list of recommended songs you can click to play. In FM mode, the app runs like a continuous radio broadcast: an AI generates DJ commentary and transitions, which are converted to speech and woven together with the music automatically. You can also ask the app about the current song, its artist, release details, or musical style. To personalize the station to your tastes, you create a local set of text files describing your music preferences, daily routines, mood rules, and seed playlists. The app reads these files to shape what it recommends and how the DJ commentary sounds. Setup requires a Claude CLI installation, a Fish Audio API key and voice ID, and a running local instance of a third-party NetEase Cloud Music API. Playing songs that require a paid NetEase account requires supplying a browser cookie from a logged-in session. The project includes a disclaimer stating it is intended for personal learning and experimentation only, and is not meant for bypassing content restrictions.

Copy-paste prompts

Prompt 1
Help me set up the Claude CLI and Fish Audio API key needed to run this AI radio app.
Prompt 2
Explain how to write the local preference files that shape this DJ's song recommendations.
Prompt 3
Walk me through the difference between Chat mode and FM mode in this app.
Prompt 4
Show me how to get a working local NetEase Cloud Music API instance running for this project.

Frequently asked questions

What is airfm?

A locally-run AI personal radio station that chats about music, recommends songs, and generates AI-voiced DJ commentary between tracks.

What language is airfm written in?

Mainly TypeScript. The stack also includes TypeScript, Claude, Fish Audio.

What license does airfm use?

No license information is stated in the README.

How hard is airfm to set up?

Setup difficulty is rated hard, with roughly 1h+ to a first successful run.

Who is airfm for?

Mainly vibe coder.

Open on GitHub → Ask about another repo

This repo across BitVibe Labs

Verify against the repo before relying on details.