Analysis of the impact of machine learning algorithms on the quality of generated sounds

Music generation using broadly understood AI is an evolving field with many challenges and opportunities. This thesis explores the use of generative adversarial networks for this endeavour, focusing on and comparing variety of different solutions that are already developed. Various architectures were tested and evaluated, in order to find the most effective approach to generating music. The results demonstrate that, although there are many solutions that can generate music that is both coherent and creative, there is still place for improvement in terms of model stability and created music quality. This work contributes to the understanding of generative adversarial networks in music generation and provides a foundation for future research in this area.