本論文提出一個多頻帶線性預估倒頻譜係數(multi-band linear predictive cepstral coefficients)的語音情緒特徵，利用離散小波轉換將訊號分解至多個子頻帶，對全頻帶和每個頻帶萃取出線性預估編碼係數，同時分析不同參數多頻帶線性預估倒頻譜係數，最後決定以分解2層、10階線性預估編碼係數和縮短取樣比例為8的做為參數。並且結合音高和能量曲線特徵，總共有52特徵，最後藉由費雪比例選擇出32個做為7種情緒的語音情緒辨識系統特徵，其整體辨識率達到90%。
最後本論文比較三種不同的類神經網路辨識器(多層感知機、放射基底函數網路和多維矩形複合式神經網路)。在整體資料集辨識率，多層感知機有90% 以上的最佳辨識率；模糊化多維矩形複合式神經網路對於訓練資料有著高達百分百的辨識結果；最後放射基底函數網路在測試資料集有68% 的辨識率。

This thesis presents a multi-band linear predictive cepstral coefficients (MBLPCC) feature for the emotional speech recognition system. Base on discrete wavelet transform (DWT), the emotional speech is decomposed into various frequency subband, and LPCC of the lower frequency subband for each decomposition process are calculated.
Furthermore, we analyze the different parameters of MBLPCC, and then decide to decompose two times, 10 LPCC coefficients and the downsampling ratio of eight as the parameters. We also combine MBLPCC with pitch and energy curve features, a total of 52 features, and choose 32 features by Fisher’s ratio for the seven kinds of emotion of emotional speech recognition system, and achieves the recognition rate of 68%.
Finally, we compare three different artificial neural networks (ANN) recognizer, multilayer perceptrons (MLP), radial basis function networks (RBF) and fuzzy hyperrectangular composite neutral networks (FHRCNN). In the recognition rate of overall data set, MLP achieved the best rate of over 90%. FHRCNN with training data set achieves recognition result of up to 100%. Finally, RBFN with testing data set achieves the recognition rate of 68%.

[1] A. Austermann, N. Esau, L. Kleinjohann, and B. Kleinjohann, “Fuzzy emotion recognition in natural speech dialogue,” Robot and Human Interactive Communication, pp. 317-322, Aug. 2005.
[2] R. Banse and K. R. Scherer, “Acoustic profiles in vocal emotion expression,” Personality and Social Psychology, vol. 70, no. 3, pp.614-636, 1996.
[3] M. W. Bhatti, Y. Wang, and L. Guan, “A neural network approach for human emotion recognition in speech,” in Proc. of the International Symposium on Circuits and System, vol.2, pp. II-181-4, May 2004.
[4] F. Burkhardt, A. Paeschke, M. Rolfes, W. Sendlmeier, and B. Weiss, “A Database of German Emotional Speech,” in Proc. of the INTERSPEECH, p.1517-1520, 2005.
[5] R. R. Cornelius, “Theoretical approaches to emotion,” in Proc. of the ISCA Workshop on Speech and Emotion, pp. 3-10, 2000.
[6] R. Cowie, E. Douglas-Cowie, B. Apolloni, J.Taypor, A. Romano, and W. Fellenz, “What a neural net needs to know about emotion words,” in Proc. of the 3rd World Multiconference on Circuits, Systems, Comms. and Computers, pp. 109-114, July, 1999.
[7] R. Cristi, Modern digital signal processing, Pacific Grove, CA, USA:Brook/Cole-Thomson Learning, 2004.
[8] F. Dellaert, T. Polzin, and A. Waibel, “Recognizing emotion in speech,” in Proc. ICSLP, pp. 1970-1973, 1996.
[9] R. O. Duda and P.E.Hart, Pattern Classification and Scene Analysis, New York：Wiley, 1973.
 
[10] F. Fragopanagos and J.G. Taylor, “Emotion recognition in human-computer interaction,” Neural Networks, vol. 18, pp. 389-405, 2005.
[11] H. H. Kyung, E. H. Kim, Y. K. Kwak, “Improvement of emotion recognition by Bayesian classifier using non-zero-pitch concept,” in IEEE International Workshop on Robot and Human Interactive Communication, pp. 312-316, Aug. 2005.
[12] D. Liu, et al., "A Structure Optimization Method Based on Fisher Ratio," in 3rd International Conf. on Natural Computation, vol. 1, pp. 54-58, 2007.
[13] I. R. Murray and J. L. Arnott, “Toward a simulation of emotion in synthetic speech: A review of the literature on human vocal emotion,” J. of the Acoustical Society of America, pp. 1097-1108, 1993.
[14] J. Nicholson, K. Takahashi, and R. Nakatsu, “Emotion recognition in speech using neural networks,” in 6th International Conf. on Neural Information Proc., vol. 2, pp. 495-501, Nov. 1999.
[15] T. L. Nwe, F. S. Wei, and L. C DE Silva, “Speech based emotion classification,” in Proc. of IEEE Region 10 International Conf. on Electrical and Electronic Technology, vol. 1, pp. 297-301, Aug 2001.
[16] A. A. Razak, R. Komiya, and M. I. Z. Abidin, “Comparison between fuzzy and NN method for speech emotion recognition,” in 3rd International Conf. on Information Technology and Applications, vol. 1, pp. 297-302, July 2005.
[17] J. Rong, Y. P. Chen, M. Chowdhury, and G. Li, “Acoustic features extraction for emotion recognition,” in 6th IEEE/ACIS International Conf. on Computer and Information Science, pp. 419-424, 2007.
[18] J. Sato and S. Morishima, “Emotion modeling in speech production using emotion space”, in 5th IEEE International Workshop on Robot and Human Communication, pp. 472-477, Nov. 1996.
[19] K. R. Scherer, “Vocal communication of emotion: A review of research paradigms,” Speech Communication, vol. 40, pp. 227-256, 2003.
[20] B. Schuller, M. Lang, and G. Rigoll, “Automatic emotion recognition by the speech signal,” in 6th World Multiconference on Systemics, Cybermetics and Informatics, pp. 367-372, 2002.
[21] B. Schuller, M. Lang, and G. Rigoll, “Hidden Markov model-based speech emotion recognition,” in Proc. of the IEEE ICASSP Conf., pp. 1-4, April 2003.
[22] B. Schuller, M. Lang, and G. Rigoll, “Speech emotion recognition combining acoustic features and linguistic information in a hybrid support vector machine-belief network architecture,” in IEEE International Conf. on Acoustics, Speech, and Signal Proc., vol. 1, pp. I-577-80, May 2004.
[23] M. M. Sondhi, “New methods of pitch extraction,” IEEE Trans. on Audio and Electroacoust, vol. 16, pp. 262-266, 1968.
[24] M. C. Su, C. -W. Liu, and S. -S. Tsay, “Neural-network-based Fuzzy Model and its Application to Transient Stability Prediction in Power Systems,” IEEE Trans. Systems, Man, and Cybernetics, vol. 29, no. 1, pp. 149-157, 1999.
[25] M. C. Su, "Use of Neural Networks as Medical Diagnosis Expert Systems," Computers in Biology and Medicine, Vol. 24, No. 6, pp. 419-429, 1994.
[26] K. Truong and D. Van Leeuwen, “An ‘open-set’ detection evaluation methodology for automatic emotion recognition in speech,” in Workshop on Paralinguistic Speech – between models and data, pp. 5-10, 2007
[27] D. Ververidis and C. Kotropoulos, “A State of the Art Review on Emotional Speech Databases,” in Proc. 1st Richmedia Conf., pp. 109-119, Oct. 2003.
 
[28] D. Ververidis and C. Kotropoulos, “Emotional speech recognition: Resources, features, and methods,” Speech Communication, vol. 48, no. 9, pp. 1162-1181, Jan. 2006.
[29] T. Vogt and E. André, “Improving automatic emotion recognition from speech via gender differentiation,” in Proc. Language Resources and Evaluation Conference, 2006.
[30] C. M. Whissell, “The dictionary of affect in language,” in Emotion: Theory, Research, and Experience, Robert Plutchik and Henry Kellerman, Ed. Academic Press, pp. 113-131.
[31] Z. Xiao, E. Dellandrea, W. Dou, and L. Chen, “Hierarchical classification of emotional speech,” IEEE Transactions on Multimedia, submitted for publication, 2007.
[32] T. Yamada, H. Hashimoto, and N. Tosa, “Pattern recognition of emotion with neural network,” in 21st International Conf. on Industrial Electronics, Control, and Instrumentation, vol. 1, pp. 183-187 Nov. 1995.
[33] Audio Signal Processing and Recognition. [Online]. Available:
http://neural.cs.nthu.edu.tw/jang/books/audioSignalProcessing/
June 22, 2009 [date accessed]
[34] Berlin Database of Emotional Speech. [Online]. Available: http://www.expressive-speech.net/emodb, June 22, 2009 [date accessed]
[35] 求是科技，數位影像處理技術大全，文魁資訊，台北市，民國九十七年。
[36] 蘇木春、張孝德，機器學習：類神經網路、模糊系統以及基因演算法則，修訂二板，全華科技圖書公司，台北，民國九十五年。
[37] 陳萬城，「雜訊環境下強健性語者辨識的新方法」，博士論文，電機工程學系，淡江大學，民國九十八年一月。