Reverse Engineering Shazam

Overview

This repository contains a reverse engineering of the Shazam app, which is used for music recognition. The goal of this project is to understand how the app works and to create a similar app using Python.

Requirements

To run the code, you need to create a virtual environment and install the required packages. You can do this by running the following command in the backend directory:

pip install -r requirements.txt

Usage

There is a separate Drive link, containing a pre-built database of fingerprints. You can use this database to identify songs without having to build your own database. The drive can be found here.

It is recommended to use the pre-built database, as building your own database can take a long time. However, if you want to build your own database, you can do so by following the instructions at the end of this README.

CLI

Identify

To identify a song using the CLI, you can use the following command:

python main.py identify <path_to_audio_file> --dir <path_to_database_directory>

If you're using the Google Drive version of the code, you can use the following command:

python main.py identify <path_to_audio_file> --dir ../data/database/

There are optional arguments you can use with the identify command:

• --hash-method - Hash method to use for fingerprinting (default: 'both', options: 'v1', 'v2', 'both')
• --duration - Duration of the audio file to use for fingerprinting (default: 30 seconds)
• --threshold - Threshold for the fingerprinting algorithm (default: 0.05)
• --workers - Number of workers to use for fingerprinting (default: 4)

Build Database

To build your own database, you can use the following command:

python main.py fingerprint --dir <path_to_song_directory> --db <path_to_database_directory>

There are optional arguments you can use with the fingerprint command:

• --workers - Number of workers to use for building the database (default: 1)
• --songs-per-db - Number of songs to include in each database file (default: 25)
• --hash-method - Hash method to use for fingerprinting (default: 'both', options: 'v1', 'v2', 'both')
• --chunk_size - How many seconds to chunk the audio file into (default: 30)

REST API - Inbuilt Website

To run the app, you need to have Python and Node installed. You can run the app on the Front by executing the following command in the backend directory:

python api.py

In the frontend directory, you can run the NextJS app by executing the following command (Note: You need to do this in a separate terminal instance):

npm run dev

Visit http://localhost:3000 to access the frontend of the app.

REST API - Custom Website

You can also access the REST API using a custom website. To do this, you need to run the Flask server in the backend directory:

python api.py

Then, you can access the API using the following endpoints:

• http://localhost:8000/api/identify - This endpoint is used to recognise music. It accepts a POST request with a file upload. The file should be in WAV format.

Notes

If you choose to build your own database, you can do the following to speed up downloading the songs necessary for the database:

1. Create a list of artist albums from Spotify in the following format: The Weeknd - Starboy - 2ODvWsOgouMbaA5xf0RkJe/Artist Name - Album Name - Spotify ID
2. Open downloader.py, and edit lines 7 and 8 to include the path to your database and the list of albums you want to download (step 1).
3. Install spotdl at SpotDL GitHub, and ensure you change the config file to save the file as format WAV.
4. You can then run the downloader script using the following command: python downloader.py.
5. Once the songs are downloaded, you can run the fingerprinting command to build your database.

If you then want to use the frontend, run the metadata.py script to create the metadata file. Once the file is created, go to the frontend directory, and run the following commands:

cd scripts
node generate-songs-data.js

This will update the song data in the frontend, allowing you to view which songs are identifiable.

Update the config.py file to include all the new paths.