Song Transposer

A full-stack web application that downloads audio from YouTube, detects its musical key, and generates a pitch-shifted (transposed) version of the audio in a new key. The project combines a React frontend, a FastAPI backend, and audio processing techniques using Python DSP libraries.

Features

• YouTube audio download using yt-dlp
• Automatic musical key detection using a Krumhansl–Schmuckler–based algorithm
• Pitch-shifting / audio transposition using librosa
• Simple UI for:
  • entering a YouTube url
  • detecting key
  • selecting a new key
  • downloading the transposed audio
• Temporary storage of downloaded and processed audio files
• CORS-enabled FastAPI server
• Clean modular separation of services and UI

Tech Stack

Frontend

• React (Vite)
• Axios

Backend

• FastAPI
• Python 3.12
• yt-dlp
• librosa, numpy, scipy

Project Structure

Song_Transposer/
│
├── backend/
│   ├── app/
│   │   ├── api/
│   │   │   ├── routes.py
│   │   ├── services/
│   │   │   ├── key_detection.py
│   │   │   ├── transposer.py
│   │   ├── utils/
│   │   │   ├── youtube.py
│   │   ├── main.py
│   ├── requirements.txt
│
├── frontend/
│   ├── src/
│   │   ├── AudioTransposer.jsx
│   │   ├── App.jsx
│   │   ├── main.jsx
│   ├── package.json
│
└── README.md

Backend Setup

1. Create and activate a virtual environment

cd backend
python3 -m venv venv
source venv/bin/activate

Windows

venv\Script\activate

2. Install backend dependencies

pip install -r requirements.txt

3. Install ffmpeg macOS

brew install ffmpeg

Ubuntu / Debian

sudp apt install ffmpeg

Windows Download from: https://ffmpeg.org/download.html and add to PATH. 4. Start the FastAPI server

uvicorn app.main:app --reload

The backend will be available at: http://127.0.0.1:8000

Frontend Setup

1. Install dependencies

cd frontend
npm install

2. Start development server

npm run dev

The frontend will be available at: http://127.0.0.1:5173

How to Use

• Open the frontend in your browser.
• Enter any YouTube video URL containing music.
• Click Download Audio.
• Click Analyse Key to detect the musical key.
• Choose your target key from the dropdown.
• Click Transpose to generate a pitch-shifted version.
• Download the processed audio.

Notes

• Audio files are stored temporarily in a system temp directory (e.g., /tmp/ytdl_audio on macOS/Linux; on Windows a platform-appropriate temp directory is used).
• The .gitignore excludes virtual environments, node_modules, caches, and temporary audio files.
• This project is cross-platform (macOS, Linux, Windows), but Windows users must install ffmpeg manually and activate the venv using venv\Scripts\activate