Circuits, Systems, and Signal Processing

, Volume 37, Issue 1, pp 232–254 | Cite as

Significance of Higher-Order Spectral Analysis in Infant Cry Classification

  • Anshu Chittora
  • Hemant A. Patil


In this paper, higher-order spectral analysis is applied to infant cry signals for classification of normal infant cries from pathological infant cries. From the family of higher-order spectra, bispectrum is considered for the proposed task. Bispectrum is the Fourier transform of the third-order cumulant function. To extract features from the bispectrum, application of higher-order singular value decomposition theorem is proposed. Experimental results show the average classification accuracy of \({82.44} \pm {4.03}{ \%}\) and Matthew’s correlation coefficient (MCC) of 0.62 with proposed bispectrum features. In all of the experiments reported in this paper, support vector machine with radial basis function kernel is used as the pattern classifier. Performance of the proposed features is also compared with the state-of-the-art methods such as linear frequency cepstral coefficients, Mel frequency cepstral coefficients, perceptual linear prediction coefficients, linear prediction coefficients, linear prediction cepstral coefficients and perceptual linear prediction cepstral coefficients, and is found to be better than that given by these feature sets. The proposed bispectrum-based features are shown to be robust under signal degradation or noisy conditions at various SNR levels. Performance in the presence of noise is compared with the state-of-the-art spectral feature sets using MCC scores. In addition, effectiveness of cryunit segmentation in normal and pathological infant cry classification task is reported.


Bispectrum Cumulants Higher-order singular value decomposition theorem Higher-order spectral analysis Support vector machine classifier 



The authors would like to thank authorities of DA-IICT, Gandhinagar, Department of Electronics and Information Technology (DeitY), New Delhi, and Department of Science and Technology (DST), New Delhi, for providing necessary resources to carry out this research work.


  1. 1.
    H.E. Baeck, M.N. Souza, Study of acoustic features of newborn cries that correlate with the context, in Proceedings of the 23rd Annual International Conference of the IEEE Engineering in Medicine and Biology Society, vol. 3, pp. 2174–2177 (2001)Google Scholar
  2. 2.
    S.E. Barajas-Montiel, C.A. Reyes-Garcia, Identifying pain and hunger in infant cry with classifiers ensembles, in International Conference on Intelligent Agents, Web Technologies and Internet Commerce, Vienna, Austria, vol. 2, pp. 770–775 (2005)Google Scholar
  3. 3.
    A. Chittora, Crying for a reason: a signal processing based approach for infant cry analysis, Ph.D. Thesis, Dhirubhai Ambani Institute of Information and Communication Technology, Gandhinagar, Gujarat, India (2017)Google Scholar
  4. 4.
    A. Chittora, H.A. Patil, Analysis of normal and pathological infant cries using bispectrum features derived using HOSVD, in International Conference on BioSignal Analysis, Processing and Systems (ICBAPS), Kuala Lumpur, Malaysia, pp. 151–155 (2015)Google Scholar
  5. 5.
    A. Chittora, H.A. Patil, Classification of normal and pathological infant cries using bispectrum features, in 23rd European Signal Processing Conference (EUSIPCO), Nice, France, pp. 639–643 (2015)Google Scholar
  6. 6.
    A. Chittora, H.A. Patil, Data collection and corpus design for analysis of normal and pathological infant cry, in Oriental COCOSDA held jointly with International Conference on Asian Spoken Language Research and Evaluation (O-COCOSDA/CASLRE), vol. 2, pp. 1–6 (2013)Google Scholar
  7. 7.
    Elemetrics kay massachusetts eye & ear infirmary voice disorder database (MEEI Database): elemetrics disordered voice database (version 1.03) (1994)Google Scholar
  8. 8.
    R. Gupta et al., Pathological speech processing: state-of-the-art, current challenges and future directions, in International Conference in Acoustics, Speech and Signal Processing (ICASSP), Shanghai, China, vol. 1, pp. 6470–6474 (2016)Google Scholar
  9. 9.
    M. Hariharan, L.S. Chee, S. Yaacob, Analysis of infant cry through weighted linear prediction cepstral coefficients and probabilistic neural networks. J. Med. Syst. 36(3), 1309–1315 (2010)CrossRefGoogle Scholar
  10. 10.
    Higher order spectral analysis toolbox (online), Accessed on 26 Sept 2016
  11. 11.
  12. 12.
    L.D. Lathauwer, B.D. Moor, J. Vandewalle, A multilinear singular value decomposition. SIAM J. Matrix Anal. Appl. 21(4), 1253–1278 (2000)MathSciNetCrossRefzbMATHGoogle Scholar
  13. 13.
    D. Lederman et al., Classification of cries of infants with cleft-palate using parallel hidden Markov models. Med. Biol. Eng. Comput. 46(10), 965–975 (2008)CrossRefGoogle Scholar
  14. 14.
    S. Li, Y. Liu, Feature extraction of lung sounds based on bispectrum analysis, in Third International Symposium on Information Processing (ISIP), Qingdao, China, pp. 393–397 (2010)Google Scholar
  15. 15.
    B.W. Matthews, Comparison of the predicted and observed secondary structure of T4 phage lysozyme. Biochim. Biophys. Acta 405(2), 442–451 (1975)CrossRefGoogle Scholar
  16. 16.
    R. Nicollas et al., The very first cry: a multidisciplinary approach toward a model. Ann. Otol. Rhinol. Laryngol. 121(12), 821–826 (2012)CrossRefGoogle Scholar
  17. 17.
    C.L. Nikias, J.M. Mendel, Signal processing with higher-order spectra. IEEE Signal Process. Mag. 10(3), 10–37 (1993)CrossRefGoogle Scholar
  18. 18.
    C.L. Nikias, R.R. Mysore, Bispectrum estimation: a digital signal processing framework. Proc. IEEE 75(7), 869–891 (1987)CrossRefGoogle Scholar
  19. 19.
    NOISEX-92. Accessed on 26 Sept 2016 (online)
  20. 20.
    P.F. Ostwald, M. Thomas, The communicative and diagnostic significance of infant sounds, in Infant Crying: Theoretical and Research Perspective, ed. by C.F.Z. Boukydis, B.M. Lester (Plenum Publishing Corporation, New York, 1985)Google Scholar
  21. 21.
    H.A. Patil, Cry baby: using spectrographic analysis to assess neonatal health from an infant’s cry, in Advances in Speech Recognition, Mobile Environments, Call Centres and Clinics Ed. Neustein Amy, Clinics, ed. by N. Amy (Springer, 2010), pp. 323–348Google Scholar
  22. 22.
    M. Petroni et al., A robust and accurate cross-correlation-based fundamental frequency (\(F_{0}\)) determination method for the improved analysis of infant cries, in IEEE 17th Annual Conference on engineering in Medicine and Biology Society, Montreal, vol. 2, pp. 975–976 (1995)Google Scholar
  23. 23.
    M. Petroni et al., Classification of infant cry vocalizations using artificial neural networks, in International Conference on Acoustics, Speech and Signal Processing (ICASSP), Detroit, vol. 5, pp. 3475–3478 (1995)Google Scholar
  24. 24.
    M. Petroni, M.E. Malowany, C.C. Johnston, B.J. Stevens, A crosscorrelation based method for improved visualization of infant cry vocalizations, in Canadian Conference on Electrical and Computer Engineering, Halifax, Canada, pp. 25–28 (1994)Google Scholar
  25. 25.
    X. Qiaobing, Automatic infant cry analysis and recognition, Ph.D. Thesis, Department of Electrical Engineering, University of British Columbia, Vancouver, Canada (1993)Google Scholar
  26. 26.
    M.A. Ruiz, C.A. Reyes, L.C. Altamirano, On the implementation of a method for automatic detection of infant cry units. Proc. Eng. 35(1), 217–222 (2012)CrossRefGoogle Scholar
  27. 27.
    J. Soltis, The signal functions of infant crying. J. Behav. Brain Sci. 27(4), 443–490 (2004)Google Scholar
  28. 28.
    C.A. Stifter, Crying behaviour and its impact on psychosocial child development, in Encyclopedia on early childhood development, Centre of Excellence for Early Childhood Development, pp. 1–5 (2005)Google Scholar
  29. 29.
    Q. Xie, R.K. Ward, C.A. Laszlo, Automatic detection of infant’s level of distress from the cry signals. IEEE Trans. Speech Audio Process. 4(4), 253–265 (1996)CrossRefGoogle Scholar
  30. 30.
    Z.R. Yang et al., RONN: the bio-basis function neural network technique applied to the detection of natively disordered regions in protein. Bioinformatics 21(16), 3369–3376 (2005)CrossRefGoogle Scholar
  31. 31.
    H. Yu et al., Feature extraction and classification based on bispectrum for underwater targets, in International Conference on Intelligent System Design and Engineering Application (ISDEA), Changsha, vol. 1, pp. 742–745 (2010)Google Scholar
  32. 32.
    A. Zabidi et al., Mel-frequency cepstrum coefficient analysis of infant cry with hypothyroidism, in 5th International Colloquium on Signal Processing and Its Applications (CSPA), Kuala Lumpur, Malaysia, pp. 204–208 (2009)Google Scholar

Copyright information

© Springer Science+Business Media New York 2017

Authors and Affiliations

  1. 1.Dhirubhai Ambani Institute of Information and Communication TechnologyGandhinagarIndia

Personalised recommendations