Chordnova: Real-Time Chord Detection Web App

Abstract

ChordNova is a browser-based intelligent web application designed for real-time chord recognition using advanced audio signal processing and machine learning. It captures live or uploaded audio through the Web Audio API and applies Constant-Q Transform (CQT) and Short-Time Fourier Transform (STFT) techniques to extract harmonic features. These are analyzed using Convolutional Neural Networks (CNN) and Recurrent Neural Networks (RNN) to classify chords, keys, and scales with high accuracy. The system provides interactive chord diagrams for guitar and piano and integrates practice and visualization modes to support learners and educators. Built on TensorFlow.js and React.js, ChordNova enables instant, crossplatform chord detection without installation. It aims to enhance music education, facilitate self-learning, and bridge the gap between auditory recognition and instrumental performance.