It is common to listen to songs that match one's mood. Thus, an AI music recommendation system that is aware of the user's emotions is likely to provide a superior user experience to one that is unaware. In this paper, we present an emotion-aware music recommendation system. Multiple models are discussed and evaluated for affect identification from a live image of the user. We propose two models: DRViT, which applies dynamic routing to vision transformers, and InvNet50, which uses involution. All considered models are trained and evaluated on the AffectNet dataset. Each model outputs the user's estimated valence and arousal under the circumplex model of affect. These values are compared to the valence and arousal values for songs in a Spotify dataset, and the top-five closest-matching songs are presented to the user. Experimental results of the models and user testing are presented.

An extension of the simply-typed lambda calculus is presented which contains both well-structured inductive and coinductive types, and which also identifies a class of types for which general recursion is possible. The motivations for this work are certain natural constructions in category theory, in particular the notion of an algebraically bounded functor, due to Freyd. We propose that this is a particularly elegant core language in which to work with recursive objects, since the potential for general recursion is contained in a single operator which interacts well with the facilities for bounded iteration and coiteration.