Description:
This thesis is part of a bigger collaborative research with the Metropolitan University of Oslo, that focuses on measuring how music affects the way an infant's brain is developed. The goal of this thesis is to build a platform for analysing musical signals and discovering knowledge with regards to the musical attributes of lullabies and songs for babies. Towards this end, audio feature extraction using traditional signal processing combined with both supervised and unsupervised deep learning models will be used. Prerequisites: (1) ML/DL background (2) python (3) pytorch (4) (optional) basics of signal processing (5) (optional) some interactive UI framework for python such as dash.