Bat Call Identification with Gaussian Process Multinomial Probit Regression and a Dynamic Time Warping Kernel

We study the problem of identifying bat species from echolocation calls in order to build automated bioacoustic monitoring algorithms. We employ the Dynamic Time Warping algorithm which has been successfully applied for bird ight calls identication and show that classication performance is superior to hand crafted call shape parameters used in previous research. This highlights that generic bioacoustic software with good classication rates can be constructed with little domain knowledge. We conduct a study with eld data of 21 bat species from the north and central Mexico using a multinomial probit regression model with Gaussian process prior and a full EP approximation of the posterior of latent function values. Results indicate high classication accuracy across almost all classes while misclassication rate across families of species is low highlighting the common evolutionary path of echolocation in bats.

[1]  Christian Dietz,et al.  A continental-scale tool for acoustic identification of European bats , 2012 .

[2]  Tom Minka,et al.  Expectation Propagation for approximate Bayesian inference , 2001, UAI.

[3]  Chih-Jen Lin,et al.  LIBSVM: A library for support vector machines , 2011, TIST.

[4]  M. Obrist.  Flexible bat echolocation: the influence of individual, habitat and conspecifics on sonar signal design , 1995, Behavioral Ecology and Sociobiology.

[5]  E. Britzke,et al.  VARIATION IN SEARCH-PHASE CALLS OF BATS , 2001 .

[6]  Neil D. Lawrence,et al.  Efficient Nonparametric Bayesian Modelling with Sparse Gaussian Process Approximations , 2006 .

[7]  Aki Vehtari,et al.  Nested Expectation Propagation for Gaussian Process Classification with a Multinomial Probit Likelihood , 2012, 1207.3649.

[8]  M. Fenton,et al.  Recognition of Species of Insectivorous Bats by Their Echolocation Calls , 1981 .

[9]  S. Chiba,et al.  Dynamic programming algorithm optimization for spoken word recognition , 1978 .

[10]  S. Parsons,et al.  Acoustic identification of twelve species of echolocating bat by discriminant function analysis and artificial neural networks. , 2000, The Journal of experimental biology.

[11]  Mark Girolami,et al.  Variational Bayesian Multinomial Probit Regression with Gaussian Process Priors , 2006, Neural Computation.

[12]  Aki Vehtari,et al.  Nested expectation propagation for Gaussian process classification , 2013, J. Mach. Learn. Res..

[13]  H. Schnitzler,et al.  From spatial orientation to food acquisition in echolocating bats , 2003 .

[14]  Christopher K. I. Williams,et al.  Gaussian Processes for Machine Learning (Adaptive Computation and Machine Learning) , 2005 .

[15]  Bing-Yu Sun,et al.  A Study on the Dynamic Time Warping in Kernel Machines , 2007, 2007 Third International IEEE Conference on Signal-Image Technologies and Internet-Based System.

[16]  Carl E. Rasmussen,et al.  Assessing Approximate Inference for Binary Gaussian Process Classification , 2005, J. Mach. Learn. Res..

[17]  Theodoros Damoulas,et al.  Bayesian Classification of Flight Calls with a Novel Dynamic Time Warping Kernel , 2010, 2010 Ninth International Conference on Machine Learning and Applications.

[18]  Daryl E. Wilson,et al.  Mammal Species of the World: A Taxonomic and Geographic Reference , 1993 .

[19]  Mingjun Zhong,et al.  Data Integration for Classification Problems Employing Gaussian Process Priors , 2006, NIPS.

[20]  Ingemar Ahlén,et al.  Use of ultrasound detectors for bat studies in Europe: experiences from field identification, surveys, and monitoring , 1999 .

[21]  M. Willig,et al.  Carpe noctem: the importance of bats as bioindicators , 2009 .

[22]  Gareth Jones,et al.  The evolution of echolocation in bats. , 2006, Trends in ecology & evolution.