Compensating for the effects of site and equipment variation on delphinid species identification from their echolocation clicks

TitleCompensating for the effects of site and equipment variation on delphinid species identification from their echolocation clicks
Publication TypeJournal Article
Year of Publication2015
AuthorsRoch M.A, Stinner-Sloan J., Baumann-Pickering S., Wiggins SM
JournalJournal of the Acoustical Society of America
Volume137
Pagination22-29
Date Published2015/01
Type of ArticleArticle
ISBN Number0001-4966
Accession NumberWOS:000348369000023
Keywordsclassification; score normalization; speaker recognition; speech; verification
Abstract

A concern for applications of machine learning techniques to bioacoustics is whether or not classifiers learn the categories for which they were trained. Unfortunately, information such as characteristics of specific recording equipment or noise environments can also be learned. This question is examined in the context of identifying delphinid species by their echolocation clicks. To reduce the ambiguity between species classification performance and other confounding factors, species whose clicks can be readily distinguished were used in this study: Pacific white-sided and Risso's dolphins. A subset of data from autonomous acoustic recorders located at seven sites in the Southern California Bight collected between 2006 and 2012 was selected. Cepstral-based features were extracted for each echolocation click and Gaussian mixture models were used to classify groups of 100 clicks. One hundred Monte-Carlo three-fold experiments were conducted to examine classification performance where fold composition was determined by acoustic encounter, recorder characteristics, or recording site. The error rate increased from 6.1% when grouped by acoustic encounter to 18.1%, 46.2%, and 33.2% for grouping by equipment, equipment category, and site, respectively. A noise compensation technique reduced error for these grouping schemes to 2.7%, 4.4%, 6.7%, and 11.4%, respectively, a reduction in error rate of 56%-86%. (C) 2015 Acoustical Society of America.

DOI10.1121/1.4904507
Integrated Research Themes: 
Student Publication: 
No
sharknado