Document Type
Article
Language
eng
Format of Original
19 p.
Publication Date
11-2009
Publisher
MDPI
Source Publication
Algorithms
Source ISSN
1999-4893
Original Item ID
doi: 10.3390/a2041410
Abstract
Using Hidden Markov Models (HMMs) as a recognition framework for automatic classification of animal vocalizations has a number of benefits, including the ability to handle duration variability through nonlinear time alignment, the ability to incorporate complex language or recognition constraints, and easy extendibility to continuous recognition and detection domains. In this work, we apply HMMs to several different species and bioacoustic tasks using generalized spectral features that can be easily adjusted across species and HMM network topologies suited to each task. This experimental work includes a simple call type classification task using one HMM per vocalization for repertoire analysis of Asian elephants, a language-constrained song recognition task using syllable models as base units for ortolan bunting vocalizations, and a stress stimulus differentiation task in poultry vocalizations using a non-sequential model via a one-state HMM with Gaussian mixtures. Results show strong performance across all tasks and illustrate the flexibility of the HMM framework for a variety of species, vocalization types, and analysis tasks.
Recommended Citation
Ren, Yao; Johnson, Michael T.; Clemins, Patrick J.; Darre, Michael; Glaeser, Sharon Stuart; Osiejuk, Tomasz S.; and Out-Nyarko, Ebenezer, "A Framework for Bioacoustic Vocalization Analysis Using Hidden Markov Models" (2009). Electrical and Computer Engineering Faculty Research and Publications. 48.
https://epublications.marquette.edu/electric_fac/48
Comments
Published version. Algorithms, Vol. 2, No. 4 (November 2009): 1410-1428. DOI. © 2009 MDPI. Published under Creative Commons Attribution License 3.0.