Hybrid feature extraction from ultrasound images for a silent speech interface
LU Wenhuan1, QU Yuexin1, YANG Yalong1, WANG Jianrong2, DANG Jianwu2
1. School of Computer Software, Tianjin University, Tianjin 300350, China;
2. School of Computer Science and Technology, Tianjin University, Tianjin 300350, China
Abstract：Principal component analysis (PCA) and discrete cosine transform (DCT) are used to extract features from ultrasound images to build an ultrasound based silent speech interface. The critical information in the image is presented by using three hybrid feature extraction methods. The first method uses PCA to extract discrete wavelet transform coefficient features. The second and third methods truncate the DCT or Walsh-Hadamard transform coefficients to the appropriate number according to the energy with the truncated coefficients then used by PCA to extract the features. Tests show that this hybrid feature extraction method outperforms standalone PCA or DCT analyses. The block DCT-PCA method gives the best result among all the methods.
路文焕, 曲悦欣, 杨亚龙, 王建荣, 党建武. 无声语音接口中超声图像的混合特征提取[J]. 清华大学学报（自然科学版）, 2017, 57(11): 1159-1162,1169.
LU Wenhuan, QU Yuexin, YANG Yalong, WANG Jianrong, DANG Jianwu. Hybrid feature extraction from ultrasound images for a silent speech interface. Journal of Tsinghua University(Science and Technology), 2017, 57(11): 1159-1162,1169.
Denby B, Oussar Y, Dreyfus G, et al. Prospects for a silent speech interface using ultrasound imaging[C]//IEEE International Conference on Acoustics, Speech and Signal Processing. Piscataway, NJ, USA:IEEE Press, 2006:365-368.
[3]
Hueber T, Chollet G, Denby B, et al. Acquisition of ultrasound, video and acoustic speech data for a silent-speech interface application[J]. Proc of ISSP, 2008:365-369.
[4]
Denby B, Stone M. Speech synthesis from real time ultrasound images of the tongue[C]//IEEE International Conference on Acoustics, Speech and Signal Processing. Piscataway, NJ, USA:IEEE Press, 2004:685-688.
[5]
Hueber T, Aversano G, Chollet G, et al. Eigentongue feature extraction for an ultrasound-based silent speech interface[C]//IEEE International Conference on Acoustics, Speech and Signal Processing. Piscataway, NJ, USA:IEEE Press, 2007:1245-1248.
[6]
Cai J, Denby B, Roussel-Ragot P, et al. Recognition and real time performances of a lightweight ultrasound based silent speech interface employing a language model[C]//INTERSPEECH. Baixas, France:ISCA, 2011:1005-1008.
[7]
Hueber T, Benaroya E L, Chollet G, et al. Development of a silent speech interface driven by ultrasound and optical images of the tongue and lips[J]. Speech Communication, 2010, 52(4):288-300.
[8]
Safari M, Harandi M T, Araabi B N. A SVM-based method for face recognition using a wavelet PCA representation of faces[C]//International Conference on Image Processing. Piscataway, NJ, USA:IEEE Press, 2004:853-856.
[9]
Puyati W, Walairacht A. Efficiency improvement for unconstrained face recognition by weightening probability values of modular PCA and wavelet PCA[C]//International Conference on Advanced Communication. Piscataway, NJ, USA:IEEE Press, 2008:1449-1453.
[10]
Chitaliya N G, Trivedi A I. Feature extraction using Wavelet-PCA and neural network for application of object classification & face recognition[C]//International Conference on Computer Engineering and Applications. Piscataway, NJ, USA:IEEE Press, 2010:510-514.
[11]
Akrouf S, Sehili M A, Chakhchoukh A, et al. Face recognition using PCA and DCT[C]//Proceedings Fifth International Conference on MEMS, Nano, and Smart Systems. Los Alamitos, CA, USA:IEEE Computer Society, 2009:15-19.
[12]
Hong X, Yao H, Wan Y, et al. A PCA based visual DCT feature extraction method for lip-reading[C]//International Conference on Intelligent Information Hiding and Multimedia Signal Processing. Los Alamitos, CA, USA:IEEE Computer Society, 2006:321-326.
[13]
Hassan M, Osman I, Yahia M. Walsh-hadamard transform for facial feature extraction in face recognition[J]. Proceedings of World Academy of Science Engineering & Technolog, 2007, 1(3):1264-1268.
[14]
Young S J, Jansen J, Odell J J, et al. The HTK Hidden Markov Model Toolkit Book[M]. Cambridge:Entropic Cambridge Research Laboratory, 1995.
[15]
Yuan J, Ryant N, Liberman M, et al. Automatic phonetic segmentation using boundary models[C]//INTERSPEECH. Lyon, France:ISCA, 2013:2306-2310.
2017, Vol. 57 
