METHOD BASED ON CONVOLUTIONAL NEURAL NETWORK FOR MUSICAL CHORD RECOGNITION
Authors
Shakhova Polina
National Technical University of Ukraine "Igor Sikorsky Kyiv Polytechnic Institute", Ukraine
Kolomiiets Yevheniia
National Technical University of Ukraine "Igor Sikorsky Kyiv Polytechnic Institute", Ukraine
Volokyta Artem
National Technical University of Ukraine "Igor Sikorsky Kyiv Polytechnic Institute", Ukraine
Abstract
The article deals with the issue of Automatic Chord Recognition (ACR) using Convolutional Neural Networks (CNN). Recognition is based on the Mel-Frequency Cepstral Coefficients (MFCC) of the input audio signal. The proposed system allows to recognize 25 basic chords. Python language and TensorFlow library were used for the development.
Keywords: Automatic Chord Recognition, Convolutional Neural Networks, MFCC, TensorFlow.