Computational Intelligence to Recognize Animal Vocalization and Diagnose Animal Health Status

Conference paper
Part of the Studies in Computational Intelligence book series (SCI, volume 445)

Abstract

Without information there is no life. In the course of evolution, nature made use of nearly every physical principle to enable organisms to gain information from and about their environment, and to affect their environment. In animal realm, sound is one of the most prominent communication means, suited for long and close distances. Acoustic monitoring of farm animals may serve as an efficient management tool to enhance animal health, welfare, and farm efficiency, and in general as a useful tool in animal ethology. The final goal is a call-recognizer to identify the meaning of sounds issued by animals. Such call-recognizer must be able to recognize the meaning of calls, independent from the individual animal and a more or less noisy environment. As a probabilistic method during the learning or training phase, feature vectors from known calls are calculated. From feature vectors of calls with the same meaning reference patterns are built and stored. To recognize a call it has to be calculated in the same way. The system then determines the reference pattern that is most similar to the pattern to be recognized and outputs the meaning. Despite the vocabulary size and complexity of human speech, which is unique in animal realm, sound production and reception have several commonalities among vertebrates. This encourages to adapt methods and experiences from speech recognition in order to recognize animal vocalization and sounds. In speech recognition, double stochastic processes, such as Hidden Markov Models (HMMs), have proven to be very efficient. They were applied here to recognize the meaning of different animal calls in two studies, using utterances of cows as an example, and to diagnose pathological coughing of pigs. The results revealed that probabilistic methods like HMMs are well suited for monitoring animals by identifying the meaning of their vocalization and to diagnose their health status.

Keywords

Hide Markov Model Recognition Rate Speech Recognition Dynamic Time Warping Human Speech 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    von Bekesy, G.: Über die mechanische Frequenzanalyse in den Schnecken verschiedener Tiere. Akustische Zeitschrift 9, 3–11 (1944)Google Scholar
  2. 2.
    Deller, J.R., Hansen, J.H.L., Proakis, J.G.: Discrete-Time Processing of Speech Signals. IEEE Press, New York (2000)Google Scholar
  3. 3.
    Ferrari, S., Silva, M., Guarino, M., Berckmans, D.: Analysis of cough sound for diagnosis of respiration infections in intensive pig farming. Trans. Am. Soc. Agric. Biol. Eng. 51(3), 1051–1055 (2008)Google Scholar
  4. 4.
    Giesert, A., Balke, W., Jahns, G.: Probabilistic analysis of coughs in pigs to diagnose respiratory infections. Landbauforschung - vTI Agriculture and Forestry Research 61(3), 237–242 (2011)Google Scholar
  5. 5.
    Hauser, M.D.: The Evolution of Communication. MIT Press, Cambridge (1997)Google Scholar
  6. 6.
    Hoy, S.: Zu den Auswirkungen von Atemwegserkrankungen auf die Mast- und Fruchtbarkeitsleistungen der Schweine. Prakt Tierarzt 2, 121–127 (1994)Google Scholar
  7. 7.
    Jahns, G.: Call recognition to identify cow conditions—A call-recogniser translating calls to text. Comput. Electron. Agric. 62(1), 54–58 (2008)CrossRefGoogle Scholar
  8. 8.
    Jahns, G., Kowalczyk, W., Walter, K.: An application of sound processing techniques for determining conditions of cows. In: Proceeding of the 4th International Workshop on Systems, Signals and Image Processing, Poznan, Poland, May 28-30 (1997)Google Scholar
  9. 9.
    Körting, J.: Beitrag zur Entwicklung des Wissens von den Tierstimmen. PhD thesis, University of Gießen (1971)Google Scholar
  10. 10.
    Moreaux, B., Beerens, D., Gustin, P.: Development of a cough induction test in pigs: effects of SR 48968 and enalapril. J. Vet. Pharmacol. Ther. 22(6), 387–389 (1999)CrossRefGoogle Scholar
  11. 11.
    Murphy, K.: Hidden Markov Model (HMM) Toolbox for Matlab. (2005), http://www.cs.ubc.ca/~murphyk/Software/HMM/hmm.html (accessed on May 2, 2012)
  12. 12.
    Paulsen, K.: Das Prinzip der Stimmbildung in der Wirbeltierreihe und beim Menschen. Akademische Verlagsgesellschaft, Frankfurt am Main (1967)Google Scholar
  13. 13.
    Rabiner, L., Juang, B.: Fundamentals of Speech Recognition, Signal processing, vol. 103. Prentice Hall, Inc., Upper Saddle River (1993)Google Scholar
  14. 14.
    Rabiner, L.R.: A tutorial on Hidden Markov Models and selected applications in speech recognition. Proceedings IEEE 77(2), 257–286 (1989)CrossRefGoogle Scholar
  15. 15.
    Scheidt, A.: Mycoplasmal pneumonia. In: Proc. North Carolina Healthy Hogs Seminar, North Carolina, USA, November 2-5, North Carolina Swine Veterinary Group (1993), http://www.ncsu.edu/project/swine_extension/healthyhogs/book1993/scheidt1.html (accessed on May 2, 2012)
  16. 16.
    Tembrok, G.: Zoosemiotik. In: Nöth, W. (ed.) Handbuch der Semiotik, 2nd edn., pp. 260–272. Metzler-Verlag, Stuttgart (2000)Google Scholar
  17. 17.
    Young, S., Kershaw, D., Odell, J., Ollason, D., Valtchev, V., Woodland, P.: The HTK Book. Cambridge University Engineering Department, Cambridge (2006)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2013

Authors and Affiliations

  1. 1.TU BraunschweigWendeburgGermany

Personalised recommendations