Quantifying complex patterns of bioacoustic variation: Use of a neural network to compare killer whale (Orcinus orca) dialects

A quantitative measure of acoustic similarity is crucial to any study compa ring vocalizations of different species, social groups, or individuals. The goal of this study was to develop a method of extracting frequency contour s from recordings of pulsed vocalizations and to test a nonlinear index of acoustic similarity based on the error of an artificial neural network at c lassifying them. Since the performance of neural networks depends on the am ount of consistent variation in the training data, this technique can be us ed to assess such variation from samples of acoustic signals. The frequency contour extraction and the neural network index were tested on samples of one call type shared by nine social groups of killer whales. For comparison , call similarity was judged by three human subjects in pairwise classifica tion tasks. The results showed a significant correlation between the neural network index and the similarity ratings by the subjects. Both measures of acoustic similarity were significantly correlated with the groups' associa tion patterns, indicating that both methods of quantifying acoustic similar ity are biologically meaningful. An;index based on neural network analysis therefore represents an objective and repeatable means of measuring acousti c similarity, and allows comparison of results across studies, species, and rime; (C) 1999 Acoustical Society of America. [S0001-4966(99)01004-8].