One obstacle which children with autism spectrum disorders (ASDs) face when learning in a public-school environment is the lack of feeling included when learning. In this study, the term inclusion refers to time that children with ASDs spend in general education settings, interacting and/or engaging with neurotypical students and teachers. Inclusion can help students with ASDs improve their social skills, as well as academic achievement, mental health, and future success (Camargo et al., 2014). Since children with ASDs often have difficulties with social interaction skills, this can prevent their successful inclusion in general education placements. Music is a type of behaviorally-based intervention, which has proven to be effective in helping students develop the skills necessary to be successfully included, and because it is a type of activity which can serve as a bit of a distraction from the social aspect of the interaction, it can help children practice social skills and interact in a comfortable way. This study examines how music is used in public school settings to help foster the skills necessary for autistic children to be involved in standard school curriculums in order to allow them to receive the full benefits from learning in a general education setting. This study was conducted by reviewing past literature on the benefits of inclusion in special education, the benefits of music for children with ASDs, and the difference in efficacy of music interventions when conducted in an inclusive setting. Interviews with special education teachers, music educators, and music therapists were also conducted to address examples of the impact of music in this research area. The study found that music is beneficial in allowing more students to be included in standard school curriculums, and data showed the trend that inclusion positively affected their social and academic development.

This project seeks to motivate runners by creating an application that selectively plays music based on smartwatch metrics. This is done by analyzing metrics collected through a person’s smartwatch such as heart rate or running power and then selecting the music that best fits their workout’s intensity. This way, as the workout becomes harder for the user, increasingly motivating music is played.

Standardization is sorely lacking in the field of musical machine learning. This thesis project endeavors to contribute to this standardization by training three machine learning models on the same dataset and comparing them using the same metrics. The music-specific metrics utilized provide more relevant information for diagnosing the shortcomings of each model.