Name That Tune: Finding a Song from a Sung Query

Explore the fascinating world of Music Information Retrieval in this comprehensive lecture by Bryan Pardo from the University of Michigan. Dive into the concept of "Query by Humming" (QBH) and its application in searching vast online music catalogs. Learn about two innovative approaches for measuring similarity between database themes and sung queries: probabilistic string-alignment algorithms and hidden Markov models (HMMs). Discover the challenges and advantages of each method, including retrieval ability and processing speed. Gain insights into the broader implications of this research for auditory user interfaces and human-computer interaction. Understand how machine learning, probabilistic natural language processing, and database search techniques are applied to create more intuitive ways of finding and retrieving music. Explore the intersection of computer science, linguistics, and music technology as Pardo shares his expertise in this cutting-edge field.