A Framework for Bioacoustic Vocalization Analysis Using Hidden Markov Models
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, a...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2009-11-01
|
| Series: | Algorithms |
| Subjects: | |
| Online Access: | http://www.mdpi.com/1999-4893/2/4/1410/ |
| _version_ | 1818520538335674368 |
|---|---|
| author | Ebenezer Out-Nyarko Sharon Stuart Glaeser Michael Darre Patrick J. Clemins Michael T. Johnson Yao Ren Tomasz S. Osiejuk |
| author_facet | Ebenezer Out-Nyarko Sharon Stuart Glaeser Michael Darre Patrick J. Clemins Michael T. Johnson Yao Ren Tomasz S. Osiejuk |
| author_sort | Ebenezer Out-Nyarko |
| collection | DOAJ |
| description | 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. |
| first_indexed | 2024-12-11T01:38:50Z |
| format | Article |
| id | doaj.art-7e9aa66f08e942388221c333534409fb |
| institution | Directory Open Access Journal |
| issn | 1999-4893 |
| language | English |
| last_indexed | 2024-12-11T01:38:50Z |
| publishDate | 2009-11-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Algorithms |
| spelling | doaj.art-7e9aa66f08e942388221c333534409fb2022-12-22T01:25:06ZengMDPI AGAlgorithms1999-48932009-11-01241410142810.3390/a2041410A Framework for Bioacoustic Vocalization Analysis Using Hidden Markov ModelsEbenezer Out-NyarkoSharon Stuart GlaeserMichael DarrePatrick J. CleminsMichael T. JohnsonYao RenTomasz S. OsiejukUsing 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.http://www.mdpi.com/1999-4893/2/4/1410/Hidden Markov Model (HMM)Greenwood Frequency Cepstral Coefficients (GFCCs) |
| spellingShingle | Ebenezer Out-Nyarko Sharon Stuart Glaeser Michael Darre Patrick J. Clemins Michael T. Johnson Yao Ren Tomasz S. Osiejuk A Framework for Bioacoustic Vocalization Analysis Using Hidden Markov Models Algorithms Hidden Markov Model (HMM) Greenwood Frequency Cepstral Coefficients (GFCCs) |
| title | A Framework for Bioacoustic Vocalization Analysis Using Hidden Markov Models |
| title_full | A Framework for Bioacoustic Vocalization Analysis Using Hidden Markov Models |
| title_fullStr | A Framework for Bioacoustic Vocalization Analysis Using Hidden Markov Models |
| title_full_unstemmed | A Framework for Bioacoustic Vocalization Analysis Using Hidden Markov Models |
| title_short | A Framework for Bioacoustic Vocalization Analysis Using Hidden Markov Models |
| title_sort | framework for bioacoustic vocalization analysis using hidden markov models |
| topic | Hidden Markov Model (HMM) Greenwood Frequency Cepstral Coefficients (GFCCs) |
| url | http://www.mdpi.com/1999-4893/2/4/1410/ |
| work_keys_str_mv | AT ebenezeroutnyarko aframeworkforbioacousticvocalizationanalysisusinghiddenmarkovmodels AT sharonstuartglaeser aframeworkforbioacousticvocalizationanalysisusinghiddenmarkovmodels AT michaeldarre aframeworkforbioacousticvocalizationanalysisusinghiddenmarkovmodels AT patrickjclemins aframeworkforbioacousticvocalizationanalysisusinghiddenmarkovmodels AT michaeltjohnson aframeworkforbioacousticvocalizationanalysisusinghiddenmarkovmodels AT yaoren aframeworkforbioacousticvocalizationanalysisusinghiddenmarkovmodels AT tomaszsosiejuk aframeworkforbioacousticvocalizationanalysisusinghiddenmarkovmodels AT ebenezeroutnyarko frameworkforbioacousticvocalizationanalysisusinghiddenmarkovmodels AT sharonstuartglaeser frameworkforbioacousticvocalizationanalysisusinghiddenmarkovmodels AT michaeldarre frameworkforbioacousticvocalizationanalysisusinghiddenmarkovmodels AT patrickjclemins frameworkforbioacousticvocalizationanalysisusinghiddenmarkovmodels AT michaeltjohnson frameworkforbioacousticvocalizationanalysisusinghiddenmarkovmodels AT yaoren frameworkforbioacousticvocalizationanalysisusinghiddenmarkovmodels AT tomaszsosiejuk frameworkforbioacousticvocalizationanalysisusinghiddenmarkovmodels |