A Combined Classifier of Cry Units with New Acoustic Attributes

  • Sergio Cano
  • Israel Suaste
  • Daniel Escobedo
  • Carlos A. Reyes-García
  • Taco Ekkel
Part of the Lecture Notes in Computer Science book series (LNCS, volume 4225)


The present work proposes a combined classifier of infant cry units that links in a single structure two focuses: a threshold-based classification and ANN-based classification. The threshold-based classifier considers 4 new acoustic features:stridor, melody, voicedness, shifts, that show properly their robustness in front of alterations of the acoustics of infant cry concerned with the presence of some diseases. In order to satisfy the automatic estimation their practical implementations are also considered. The ANN-based classifier consists in a feed-forward network using the method of Scale Gradient Conjugate (MSGC) as learning algorithm and the MFCCs as input vectors to the net. Each focus or classification stage gives in the exit one indicator (FN1 and FN2) that generates to the output a decision on two classes with gradation (normal, moderately-pathologic and pathologic). The results demonstrate the potentiality of these types of combined classifiers when the advantages of each focus in particular are properly emphasized


Cry classification pattern recognition neural network 


  1. 1.
    Petroni, M., Malowany, A., Johnston, C., Stevens, B.: Identification of Pain from Infant Cry Vocalizations Using Artificial Neural Networks (ANNs). In: The International Infant Cry Research Group. Applications and Science of Artificial Neural Networks. The International Society for Optical Engineering, vol. 2492, pp. 729–738 (1995)Google Scholar
  2. 2.
    Schonweiler, R., et al.: Neuronal networks and self-organizing maps: new computer techniques in the acoustic evaluation of the infant cry. Int. Journal of Pediatric Otorhinolaryngology 38, 1–11 (1996)CrossRefGoogle Scholar
  3. 3.
    Suaste, I., Reyes-Galaviz, O., Dias, A., Reyes, C.: A Fuzzy Relational Neural Network for Pattern C. In: Sanfeliu, A., Martínez Trinidad, J.F., Carrasco Ochoa, J.A. (eds.) CIARP 2004. LNCS, vol. 3287, pp. 358–365. Springer, Heidelberg (2004)CrossRefGoogle Scholar
  4. 4.
    Alonso, L., et al.: Reconocimiento de Patrones con Redes Neuronales. In: Alonso, L. (ed.) Editorial Imprenta Catedral, pp. 37–56 (2001)Google Scholar
  5. 5.
    Wasz-Höckert, O., Lind, J., Vuorenkoski, V., Partanen, T., Valanne, E.: The Infant Cry: A Spectrographic and Auditory Analysis. Spastics International Medical Publication, Lavenham, U.K. (1968)Google Scholar
  6. 6.
    Hurlock, E.B.: Child Development. McGraw-Hill, New York (1950)Google Scholar
  7. 7.
    Zeskind, P.S., Lester, B.: Acoustic features and auditory perceptions of the Cries of Newborns with Prenatal Complications. Child development, pp. 580–589 (1978)Google Scholar
  8. 8.
    Michelsson, K.: Sound Spectrographic cry analysyis of normal and abnormal newborns. Folia Phoniatria 28, 161–173 (1982)Google Scholar
  9. 9.
    Escobedo, D., Cano, S., Coello, E., Regueiferos, L., Capdevila, L.: Rising Shift of Pitch Frequency in the Infant Cry pf Some Pathologic Cases. In: 2nd >Int. Conf. MAVEBA 2001, Firenze, Italy (2001)Google Scholar
  10. 10.
    Cano, S., et al.: The Spectral Analysis of Infant Cry: An Initial Approximation. In: Proceedings of EUROSPEECH 1995 (sponsored by ESCA & IEEE), Madrid, September 18-21 (1995)Google Scholar
  11. 11.
    Orozco, J., Reyes, C.: Extracción y Análisis de Características Acústicas del Llanto de Bebés para su Reconocimiento Automático Basado en Redes Neuronales. Tesis de Maestría, INAOE, Puebla, Mex (2002)Google Scholar
  12. 12.
    Cano, D., et al.: Análisis preliminar de los resultados de una clasificación de unidades de llanto según tres arquitecturas de redes neuronales. Memorias de TELEC 2002 (en CD’Rom). Centro de Convenciones HEREDIA (Julio 2002)Google Scholar
  13. 13.
    Moller,: A Scaled Conjugate Gradient Algorithm for Fast Supervised Learning. Neural Networks 6(4), 525–533 (1993)CrossRefGoogle Scholar
  14. 14.
    Boersma, P., Weenink, D.: Praat v. 4.0.27. Sistema para hacer fonética por computadora. University of Amsterdam. The Netherlands 1992-2002Google Scholar
  15. 15.
    The MathWoks, Inc., MATLAB®. Neural Network Toolbox. Version 1984-2000Google Scholar
  16. 16.
    Ekkel, T.: Neural Network-Based Classification of Cries from Infants Suffering from Hypoxia-Related CNS Damage. Tesis de Maestría. University of Twente. The Netherlands (2002)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2006

Authors and Affiliations

  • Sergio Cano
    • 1
  • Israel Suaste
    • 2
  • Daniel Escobedo
    • 1
  • Carlos A. Reyes-García
    • 2
  • Taco Ekkel
    • 3
  1. 1.CENPISUniversity of OrienteStgo de CubaCuba
  2. 2.INAOEPueblaMexico
  3. 3.Faculty of InformaticsUniversity of TwenteThe Netherlands

Personalised recommendations