Music is known to impact moods in positive ways and promote well-being. Musical preferences and habits have been associated with an individual’s requirement to satisfy and reinforce their psychological needs. In fact, an individual’s musical preferences and habits have been considered a mirror to who they are emotionally, socially, and cognitively. Empirical evidence is emerging that allows decoding individual traits and states based on their musical choices and engagement styles. Certain kinds of musical engagement have also been found to be associated with measures of ill-health including risk of depression and antisocial behavior, particularly in youth. In my talk I will outline research on how individual differences affect our perception and choice of music, in addition to how the latter allows us to examine and identify individual differences.

I will describe some machine learning (ML) tools that we can use for processing music and for processing the acoustics of speech. I will then present some connections between music and speech, and between speech and depression, and I will discuss how such ML tools can help us explore these connections.