Special Issue "Algorithms for Sound Localization and Sound Classification"

Quicklinks

A special issue of Algorithms (ISSN 1999-4893).

Deadline for manuscript submissions: closed (31 March 2010)

Special Issue Editor

Guest Editor
Dr. David Chesmore (Website)

Department of Electronics, University of York, Heslington, York, UK YO10 5DD
Phone: +44 (0)1904 43 2394
Fax: +44 (0)1904 43 2335
Interests: application of electronics and computing technology to biology, ecology and entomology, particularly in the areas of automated species identification and biodiversity informatics

Keywords

  • computational bioacoustics
  • sound classification
  • automated identification
  • pattern recognition
  • acoustic source separation

Published Papers (2 papers)

View options order results:
result details:
Displaying articles 1-2
Export citation of selected articles as:

Research

Open AccessArticle A Framework for Bioacoustic Vocalization Analysis Using Hidden Markov Models
Algorithms 2009, 2(4), 1410-1428; doi:10.3390/a2041410
Received: 1 September 2009 / Revised: 30 October 2009 / Accepted: 9 November 2009 / Published: 18 November 2009
Cited by 12 | PDF Full-text (239 KB) | HTML Full-text | XML Full-text
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 [...] Read more.
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. Full article
(This article belongs to the Special Issue Algorithms for Sound Localization and Sound Classification)
Open AccessArticle Classification of Sperm Whale Clicks (Physeter Macrocephalus) with Gaussian-Kernel-Based Networks
Algorithms 2009, 2(3), 1232-1247; doi:10.3390/a2031232
Received: 9 July 2009 / Revised: 31 August 2009 / Accepted: 15 September 2009 / Published: 22 September 2009
Cited by 2 | PDF Full-text (488 KB) | HTML Full-text | XML Full-text
Abstract
With the aim of classifying sperm whales, this report compares two methods that can use Gaussian functions, a radial basis function network, and support vector machines which were trained with two different approaches known as C-SVM and ν-SVM. The methods [...] Read more.
With the aim of classifying sperm whales, this report compares two methods that can use Gaussian functions, a radial basis function network, and support vector machines which were trained with two different approaches known as C-SVM and ν-SVM. The methods were tested on data recordings from seven different male sperm whales, six containing single click trains and the seventh containing a complete dive. Both types of classifiers could distinguish between the clicks of the seven different whales, but the SVM seemed to have better generalisation towards unknown data, at the cost of needing more information and slower performance. Full article
(This article belongs to the Special Issue Algorithms for Sound Localization and Sound Classification)
Figures

Journal Contact

MDPI AG
Algorithms Editorial Office
St. Alban-Anlage 66, 4052 Basel, Switzerland
algorithms@mdpi.com
Tel. +41 61 683 77 34
Fax: +41 61 302 89 18
Editorial Board
Contact Details Submit to Algorithms
Back to Top