Method for Restoring Audio Signal Using Convolutional Neural Networks

This paper describes a research of the restoring distorted audio signal possibility. Based on previously obtained results using deep machine learning methods, concept neural network to correct has been developed. On basis originally results, several new architectures were developed, focused restoring. The contains descriptions developed with theoretical substantiation possibility their application. presented tested solve problem part specifc instrument in musical composition where it was removed. testing networks are forms.