Speaker: Mauricio Michel Olvera zambrano
Data and place: November 25, 2021, at 10:30 – Hybrid
Abstract: Ambient sound detection and classification systems are highly susceptible to performance degradation when trained and tested on data facing different acoustic conditions. In practical settings, where the source and target domain data are available, but only the source domain data are labeled, we can rely on unsupervised domain adaptation strategies to reduce data mismatches and enhance performance in the target domain. In this talk, I will present domain adaptation strategies for the detection of domestic sounds as well as for the classification of acoustic scenes.
