AudioIdentifier

An algorithm for recognizing and fingerprinting audio files, implemented in C++.

Table of Contents

• General Info
• Requirements
• Screenshots
• Setup
• Project Status
• Contribution
• TODO
• Acknowledgements

General Information

AudioIdentifier is a C++ application designed for audio fingerprinting and recognition. Unlike traditional audio recognition systems, AudioIdentifier stands out by avoiding external libraries for fingerprinting, making it a self-contained solution. The project utilizes SFML for building a graphical user interface (UI), SQLite for efficient database management, and PortAudio for handling audio input.

Requirements

• SFML
• SQLite
• PortAudio
• Cmake

Screenshots

Setup

1. Clone the Repository:

   git clone [email protected]:PlungedInCode/AudioIdentifier.git
   cd AudioIdentifier

2. Build the Project:

   mkdir build && cd build
   cmake ..
   make

3. Run project

   ./AudioIdentifier 

Project Status

Project is: in progress

Contribution

Feel free to contribute to AudioIdentifier by submitting issues or pull requests.

TODO

• Optimize peak detection algorithm
• Optimize hashing and matching algorithms
• Extract metadata from .wav files
• Better UI
• Concurrency
• Expand list of supported formats

Thanks to

• Audio Fingerprinting with Python and Numpy
• abracadabra: How does Shazam work?
• Avery Wang: An Industrial-Strength Audio Search Algorithm
• Audio Identification by Meinard Müller and Patricio López-Serrano.
• SpectroMap: Peak detection algorithm for audio fingerprinting