Song Recognition Based on Audio Finger printing

In the era of digital media, the ability to accurately identify music from audio input has gained significant importance in areas such as entertainment, copyright enforcement, and media content management. This project presents a robust and efficient system for song recognition based on audio fingerprinting techniques. By capturing real-time audio through a microphone and converting it into a unique digital signature using SHA1 hashing, the system enables accurate and fast identification of songs from a reference database. The implementation leverages open-source tools and Python libraries to ensure scalability, portability, and cost-effectiveness. The methodology involves real-time audio capture using PyAudio, preprocessing and normalization with PyDub, and feature extraction for fingerprint generation. These fingerprints are then matched against a SQLite database containing known song signatures. The system is designed to handle short audio clips and operate reliably even in the presence of background noise or distortions. This enhances its usability in real-world environments, making it a practical tool for mobile apps, music streaming platforms, and media monitoring services