AUDIO FEATURES EXTRACTION FOR NEURAL NETWORKS USAGE
Authors
Kolomiiets Yevheniia
National Technical University of Ukraine "Igor Sikorsky Kyiv Polytechnic Institute", Ukraine
Shakhova Polina
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 feature extraction of an audio signal for using the result data by a neural network that identifies the elements of harmony of a musical piece. The developed system is an audio data preprocessing service and uses such algorithms as the tonal profile algorithm (KSH algorithm) and Python, Numpy, and Librosa language tools to determine key audio.
Key words: audio data, audio analysis, music analysis, Python, audio processing.
Fig.: 4. Bibl.: 5.