本論文研製一款符合高齡者需求的陪伴機器人，並賦予機器人語音情緒辨識，以達到瞭解人類情緒並且撫慰人心的機器人。本論文與孫佾微同學之論文合作完成陪伴機器人，本文負責機器人語音辨識部分。為了實現符合高齡者需求的陪伴機器人，機器人會透過聽覺理解高齡者情緒，做出安撫情緒的語音回應以及表情動作，來達到撫慰高齡者的孤獨感以及悲傷。本文以語音辨識情緒，提出了一種基於希爾伯特-黃轉換的語音情感識別方法。有別於直接分析給定的語音訊號，使用EMD提取本質模態函數（IMF）和邊際頻譜。研究發現，對應於不同情緒的IMF個數和邊際頻譜值具有區別特徵，而以之做為辨識的特徵。為了強化辨識，使用支持向量機(SVM)進行情緒分類，對Emo-DB語音情感數據庫進行了模型訓練，該數據庫包含四種類別的情緒，即生氣、快樂、傷心和一般，總共103筆語音資料，最終獲得辨識率為84%。換言之，機器人可以獲得很好的情緒辨識以及與人的互動。

A companion robot was designed and implemented in this thesis for elderly who may be alone or feel lonely at home. The robot can recognize the emotion of the elderly through voice processing and comfort their hearts by using various motions. In the emotion recognition system, a speech emotion recognition method based on the Hilbert-Huang transform (HHT) was developed. Instead of directly analyzing a speech signal, HHT applies the empirical mode decomposition (EMD) to extract the intrinsic mode function (IMF) and combines the IMF feature, the number of IMFs, and the marginal spectrum to identify various emotion features embedded in the voice. These distinctive features and the associated statistical parameters were selected as the features for emotion recognition. In order to classify different emotions, the method of support vector machine (SVM) was used for emotion classification for which the Emo-DB speech emotion database was modeled first. The database contains four categories of emotions; namely, angry, happy, sad, and general. Then, the model was used for the present robot. The results show that the final recognition rate is 84%, indicating that the robot can recognize its user’s emotion and interact with its user reasonably well.

[1] 內政部統計處https://www.moi.gov.tw/stat/index.aspx
[2] 吳文智 (2017). 以整合型科技接受模式探討高齡者對於照護型機器人之接受度. 福祉科技與服務管理學刊, 5卷(2期).
[3] K. Wada, T. Shibata, T. Musha and S. Kimura, "Robot therapy for elders affected by dementia", IEEE Engineering in Medicine and Biology Magazine, vol. 27, no. 4, pp. 53-60, 2008.
[4] http://www.parorobots.com/, 2014
[5] http://www.riken.jp/en/pr/press/2015/20150223_2, 2015
[6] C. Jayawardena, I.-H. Kuo, U. Unger, A. Igic, R. Wong, C. I. Watson, R. Q. Stafford, E. Broadbent, P. Tiwari, J. Warren, J. Sohn and B. A. MacDonald, “Deployment of a Service Robot to Help Older People,” IEEE/RSJ International Conference on Intelligent Robots and Systems, pp. 5990-5995, 2010
[7] J. Chan and G. Nejat,” A Learning-based Control Architecture for an Assistive Robot Providing Social Engagement during Cognitively Stimulating Activities,” IEEE International Conference on Robotics and Automation, pp. 3928-3933, 2011
[8] J. Woo, K. Wada and N. Kubota, “Robot Partner System for Elderly People Care by Using Sensor Network,” The Fourth IEEE RAS/EMBS International Conference on Biomedical Robotics and Biomechatronics, pp. 1329-1334, 2012
[9] D. Kumar Limbu, W. Chern Yuen and T. Hwang Jian, "Affective social interaction with CuDDler robot", 2013 6th IEEE Conference on Robotics, Automation and Mechatronics (RAM), 2013.
[10] F. Burkhardt, A. Paeschke, M. Rolfes, W. Sendlmeier, B.Weiss, “A database of german emotional speech.” In: Proc. of the 2005 INTERSPEECH, 2005

[11] S. Steidl, “Automatic classification of emotion-related user states in spontaneous children’s speech,” [Ph.D. Thesis]. Erlangen: University at Erlangen Nurberg, 2009.
[12] https://sail.usc.edu/iemocap/,2018
[13] http://nnime.ee.nthu.edu.tw/ ,2018
[14] P. Shen, C.-J.Zhou and X. Chen, “Automatic Speech Emotion Recognition Using Support Vector Machine,” Proceedings of 2011 International Conference on Electronic & Mechanical Engineering and Information Technology, pp. 621-625, 2011
[15] A. Milton, S. S. Roy and S. T. Selvi, “SVM Scheme for Speech Emotion Recognition Using MFCC Feature,” International Journal of Computer Applications, pp. 34-39, 2013
[16] W. Zhang, X.-Y. Zhang and Y. Sun, “Based on EEMD-HHT Marginal Spectrum of Speech Emotion Recognition,” 2012 International Conference on Computing, Measurement, Control and Sensor Network, pp. 91-94, 2012
[17] Z.-L. Wang, H.-F. Li and L. Ma, “HHT Based Long Term Feature Extracting Method for Speech Emotion Classification,” Audio, Language and Image Processing, pp. 276-281, 2012
[18] M. Tahon and L. Devillers, "Towards a Small Set of Robust Acoustic Features for Emotion Recognition: Challenges", IEEE/ACM Transactions on Audio, Speech, and Language Processing, vol. 24, no. 1, pp. 16-28, 2016.
[19] N. Yang, J. Yuan, Y. Zhou, I. Demirkol, Z. Duan, W. Heinzelman and M. Sturge-Apple, "Enhanced multiclass SVM with thresholding fusion for speech-based emotion classification", International Journal of Speech Technology, vol. 20, no. 1, pp. 27-41, 2016.
[20] K. Han, D. Yu and I. Tashev, "Speech Emotion Recognition Using Deep Neural Network and Extreme Learning Machine", INTERSPEECH 2014, 2014.
[21] A. Shaw and R. Kumar, " Emotion Recognition and Classification in Speech using Artificial Neural Networks", International Journal of Computer Applications, 2016.
[22] D. Toniuc and A. Groza, "Climebot: An argumentative agent for climate change", Intelligent Computer Communication and Processing (ICCP), 13th IEEE International Conference on Cluj-Napoca, Romania, 2017
[23] http://ww1.microchip.com/downloads/en/devicedoc/70135C.pdf, November, 2017
[24] https://hobbytronics.com.pk/product/lm2596-adjustable-dc-dc-step-down-power-supply-module/, November, 2017
[25] https://www.taiwaniot.com.tw/product/ga12-n20-微型金屬減速馬達, November, 2017
[26] https://www.pololu.com/product/1595, November, 2017
[27] http://akizukidenshi.com/catalog/g/gI-02001/, November, 2017
[28] https://uge-one.com/pc817-optocoupler-optoisolator-dip-ic.html, November, 2017
[29] https://www.mi.com/tw/littleaudio/, November, 2017
[30] https://www.icshop.com.tw/product_info.php/products_id/23844, November, 2017
[31] https://zh.wikipedia.org/wiki/Node.js, November, 2017
[32] Node.js. [Online]. Available: https://nodejs.org/en/.
[33] https://dialogflow.com/, December, 2017
[34] http://blog.csdn.net/dvstream/article/details/17784293, December, 2017
[35] http://www.runoob.com/nodejs/nodejs-express-framework.html, December, 2017
[36] Cordova. [Online]. Available: https://cordova.apache.org/
[37] https://blog.csdn.net/sinat_31057219/article/details/55505634, December, 2017
[38] M. Ayadi, M. S.Kamel and F. Karray, “Survey on Speech Emotion Recognition: Features, Classification Schemes, and Databases,” Pattern Recognition, pp. 572-587, 2010
[39] N. E. Huang and S P Shen, “Hilbert-Huang Transform and Its Applications,” World Scientific Pub Co Inc,2014
[40] 楊佳穎(2008)，以HHT為基礎之肺音與哮喘音辨識研究。國立台北科技大學碩士論文。
[41] P. H. Chen, C. J. Lin, and B. Schölkopf, A tutorial on ν-support vector machines, Appl. Stoch. Models. Bus. Ind. 2005, 21, 111-136.