This paper describes an algorithm designed for Microsoft's Groove music service, which serves millions of users world wide. We consider the problem of automatically generating personalized music playlists based on queries containing a ``seed'' artist and the listener's user ID. Playlist generation may be informed by a number of information sources including: user specific listening patterns, domain knowledge encoded in a taxonomy, acoustic features of audio tracks, and overall popularity of tracks and artists. The importance assigned to each of these information sources may vary depending on the specific combination of user and seed~artist.

The paper presents a method based on a variational Bayes solution for learning the parameters of a model containing a four-level hierarchy of global preferences, genres, sub-genres and artists. The proposed model further incorporates a personalization component for user-specific preferences. Empirical evaluations on both proprietary and public datasets demonstrate the effectiveness of the algorithm and showcase the contribution of each of its components.