新生的嬰兒由於其發聲器官及心智不夠成熟，無法利用語言及肢體
動作表達自己本身的生理需求及情緒 ，因此哭聲為其唯一與外界表達溝通之途徑。在早期的研究中，可以發現嬰兒的哭聲為一個強而有力的表達工具，可以藉由嬰兒哭聲判別情緒、生理疾病和生理需求。
本文提出了一個嬰兒情緒哭聲辨識系統，分辨五種的嬰兒情緒哭
聲，即： A：秤重（Scale） B：洗澡(Bath) C：飢餓(Hunger) D：打針（Injection）和E：酒精（Alcohol）。
本研究探討利用不同的語音特徵參數，在五種嬰兒哭聲的辨識情形，並使用類神經網路和K-最近鄰居法則為辨識機制。其所使用之特徵為：倒頻譜(Cepstrum)，梅爾倒頻譜(MFCC)，線性預估參數( LPC) 和線性預估導出之倒頻譜參數（LPCC）。
實驗中，藉由改變聲學特徵向量的維數，可發現使用適合的維數，將會提高辨識率，另外，實驗分析，LPCC 為最佳辨識特徵，LPCC 不同特徵維度中，以LPCC96 為最佳LPCC 維度，平均辨識率約為83.3%。

An infant’s crying is the only communication method in the first month of life.
It can be used to describe the needs of the babies or for making medical diagnoses of pathologies at very early stages of life.
This work presents the development of an automatic recognition system of infant cry, with the objective to classify five types of crying due to: scale ,bath , hunger ,injection ,and alcohol .
Four kinds of acoustic characteristics are used to classify five types of infant cry. Both artificial neural networks (ANN) and K-nearest neighbor rule(KNNR) are examined as potential classfiers.
The acoustic characteristics used to distinguish the infant cries are Cepstrum Coefficients (Cepstrum) , Mel-Frequency Cepstrum Coefficients(MFCC) , Linear Predict Coefficients ( LPC) , LPC derived cepstrum coefficients (LPCC).
Different numbers of acoustic characteristic coefficients are used to classify infant cry in the experiments. Better classification results can be obtained with a proper number of coefficients.
The best acoustic characteristic is LPCC, and the best overall accuracy obtained with LPCC96 is about 83% .

[1] M. Petroni, A. S. Malowany, C. C. Johnston, and B. J. Stevens, “A New, Robust Vocal Fundamental Frequency (F0) Determination Method for the Analysis of Infant Cries”, Proceedings of the IEEE Symposium on Computer-Based Medical Systems, Winston-Salem, NC, USA, pp.223-228, 1994.
[2] M. Petroni, A. S. Malowany, C. C. Johnston, and B. J. Stevens, “A Robust and Accurate Cross-Correlation-Based Fundamental Frequency (F0) Determination Method for the Improved Analysis of Infant Cries”, Annual International Conference of the IEEE Engineering in Medicine and Biology – Proceedings, Montreal, Can, vol. 2,
pp. 975-976, 1995.
[3] M. Petroni, A. S. Malowany, C. C. Johnston, and B. J. Stevens, “ A Crosscorrelation-Based Method for Improved Visualization of Infant Cry Vocalizations”, Canadian Conference on Electrical and Computer Engineering, Halifax, Can, vol. 2, pp. 453-456, 1994.
[4] H. E. Baeck and M. N. Souza, “Study of Acoustic Features of Newborn Cries that Correlate with the Context”, Annual International Conference of the IEEE Engineering in Medicine and Biology Society, Kyoto University, vol. 3, pp. 2174-2177, 2001.
[5] G. Jr. Varallyay, Z. Benyo, A. Illenyi, Z. Farkas and L. Kovacs, “Acoustic Analysis of The Infant Cry: Classical and New Methods”, Annual International Conference of the IEEE Engineering in Medicine and Biology – Proceedings, San Francisco, CA, United States, IEEE International Conference, vol. 1, pp. 313-316, 2004.
[6] Ada Fort and Claudia Manfredi, “Acoustic Analysis of Newborn Infant Cry Signals”, Medical Engineering & Physics, vol.20, pp. 432 - 442, 1998.
[7] A. Fort, A. Ismaelli, C. Manfredi and P. Bruscaglioni, “Parametric and Non-Parametric Estimation of Speech Formants: Application to Infant Cry”, Medical Engineering & Physics, vol.18, pp. 677-691, 1996.
[8] T. Harada, A. Saito, T. Sato and T. Mori, “Infant Behavior Recognition System Based On Pressure Distribution Image”, Proceedings - IEEE International Conference on Robotics and Automation, San Francisco, CA, USA, vol.4, pp. 4082 - 4088, 2000.
[9] A. M. Perez, A. Gutierrez, M. Sanchez, J. Remolina and O. Aguilera, “Determination of A Mathematical Indicator From the Acoustical Analysis of Primal Crying of Newborns to Evaluate Their Well-being”, Annual International Conference of the IEEE Engineering in Medicine and Biology – Proceedings, Chicago, IL, USA, vol.3,
pp. 1073 - 1075, 1997.
[10] K. Wermke, W. Mende, C. Manfredi and P. Bruscaglioni,
“Developmental Aspects of Infant’s Cry Melody and Formants”, Medical Engineering & Physics, vol.24,
pp. 501 - 514, 2002.
[11] J. O. Garcia and C. A. Reyes Garcia, “Mel-Frequency Cepstrum Coefficients Extraction from Infant Cry for Classification of Normal and Pathological Infant with Feed-Forward Neural Networks”, Proceedings of the International Joint Conference on Neural Networks, Portland, vol. 4, pp. 3140-3145, 2003.
[12] J. O. Garcia and C. A. Reyes Garcia, “Implementation and Analysis of Training Algorithms for the Classification of Infant Cry with Feed-Forward Neural Networks”, IEEE International Symposium on Intelligent Signal Processing , pp. 271-276, 2003
[13] D. Lederman, A. Cohen , E. Zmora , K. Wermke, S. Hauschildt and A. Stellzig-Eisenhauer, “On the Use of Hidden Markov Models in Infants' Cry Classification”, The 22nd Convention of Electrical and Electronics Engineers, Israel, pp. 350-352, 2002.
[14] M. Petroni, A. S. Malowany, C. C. Johnston, and B. J. Stevens, “A Comparison of Neural Network Architectures for the Classification of Three Types of Infant Cry Vocalizations”, Annual International Conference of the IEEE Engineering in Medicine and Biology - Proceedings, Montreal, Can, vol. 1, pp. 821-822, 1995.
[15] M. Petroni, A. S. Malowany, C. C. Johnston, and B. J. Stevens, “Classification of Infant Cry Vocalizations Using Artificial Neural Networks (ANNs)”, ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings, Detroit, MI, USA ,vol. 5,
pp.3475-3478, 1995.
[16] M. Petroni, A. S. Malowany, C. C. Johnston, and B. J. Stevens, “On the Use of Artificial Neural Networks (ANNs) for the Classification of Three Types of Infant Cries”, IEEE Pacific RIM Conference on Communications, Computers, and Signal Processing – Proceedings, Victoria, BC, Can, pp. 501-504, 1995.
[17] S. E. Barajas-Montiel and C. A. Reyes Garcia, “Identifying Pain and Hunger in Infant Cry with Classifiers Ensembles”, Proceedings - International Conference on Computational Intelligence for Modelling,
Control and Automation, CIMCA 2005 and International Conference on Intelligent Agents, Web Technologies and Internet, Vienna, Austria, vol. 2, pp. 770-775, 2005.
[18] P. Pal , A .N. Iyer and R. E. Yantorno, “Emotion Detection From Infant Facial Expressions and Cries”, ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing – Proceedings, Toulouse, France, vol.2, pp. II - II, 2006.
[19] J. R. Deller, J. H. L. Hansen and J. G. Proakis, “Discrete Time Processing of Speech Signals ”, Prentice Hall, New Jersey ,1987
[20] O. Wasz-Hockert, T. Partanen, V. Vuorenkoski, E. Valanne and K. Michelsson, “The Identification of Some Specific Meanings in Infant Vocalizaton ”, Experiencia, vol.20, pp. 154-156,1964
[21] A. Ismaelli, G. Rapisardi, G. Donzelli, R. Moroni and
P. Bruscaglioni, “A New Device for Computerized Infant Cry Analysis in The NICU”, Annual International Conference of the IEEE Engineering in Medicine and Biology – Proceedings, Baltimore, MD, USA, vol. 2,
pp.854-855, 1994.
[22] X. Qiaobing, R. K. Ward and C. A. Laszlo, “Determining Normal Infants' Level-of-Distress from Cry Sounds”, Canadian Conference on Electrical and Computer Engineering, Vancouver, BC, Canada, vol. 2, pp.1094-1096, 1993.
[23] C. Manfredi, V. Tocchioni, L. Bocchi and L. Kovacs, “A Robust Tool for Newborn Infant Cry Analysis”, Annual International Conference of the IEEE Engineering in Medicine and Biology – Proceedings, New York, NY, United States, pp. 509-512, 2006.
[24] C. J. Clement, F. J. Koopmans-van Beinum and L. C. W. Pols,“Acoustical Characteristics of Sound Production of Deaf and Normally Hearing Infants”, International Conference on Spoken Language Processing, ICSLP, Proceedings, Philadelphia, PA, USA, vol.3,
pp. 1073-1075, 1997.
[25]王小川，語音訊號處理，全華科技圖書公司，民國94年。
[26] 張柏雄著，“中文語音情緒之自動辨識” ，國立成功大學碩士論文，民國90年。
[27] 陳雅菁著，“類神經網路在兒童構音異常診斷上之應用” ，國立成大學碩士論文，民國84年。
[28]謝依蘭，語音訊號數位處理，松岡圖書公司，民國81年。
[29]魏渭堂，彩色圖說嬰幼兒發展保育學－學齡前兒童的照顧與發展，合計圖書出版社，民國95年。
[30]盧素碧，嬰幼兒保育，文景出版社，民國77年。
[31]Jyh-Shing Roger Jang，線上中文教材：音訊處理與辨識
Home page : http://www.cs.nthu.edu.tw/~jang.
[32]Jyh-Shing Roger Jang，線中上文教材：資料群聚與樣式辨識
Home page : http://www.cs.nthu.edu.tw/~jang
[33]張銘豐著，“以時頻域特徵參數為主之嬰兒哭聲辨識系統研製” ，逢甲大學碩士論文,民國93年。