Mixtures of probability experts for audio retrieval and indexing

Abstract

This paper describes a system for connecting nonspeech sounds and words using linked multidimensional vector spaces. An approach based on a mixture of experts learns the mapping between one space and the other. This paper describes the conversion of audio and semantic data into their respective vector spaces. Two different mixture-of-probability-expert models are trained to learn the association between acoustic queries and the corresponding semantic explanation, and vice versa. Test results are presented based on commercial sound effects CDs.