Швай, НадіяПроцик, Олексій2022-01-172022-01-172021https://ekmair.ukma.edu.ua/handle/123456789/22287The aim of this research is to analyze the current methods of audio upsampling and the analysis of the capabilities of Neural Networks as generative models in the audio upsampling domain. The task of this research is to conduct thorough experiments on the fidelity and execution time of current signal processing methods and to create and train a Neural Network, which would have a competitive comparison to original methods, while producing higher fidelity audio.enдискретизація звукуметоди збільшення роздільної здатностімагістерська роботаOther