聲學回音(Acoustic Echo)對於智慧型行動裝置使用者是主要干擾語音辨識的因素之一，本研究提出新的適應性回音消除(Adaptive Echo Cancellation , 簡稱AEC)演算法，在雙端對講的情況下，利用雙麥克風架構，將副麥克風作為主麥克風的外部語音訊號估測值，在適應性濾波器更新係數時去除外部語音訊號的影響，一方面可以解決外部語音訊號的干擾造成估測誤差提升的問題，另一方面也可使適應性濾波器更新時不隨外部語音訊號能量大小而波動，除此之外，本架構在單端發話的情況，也是能夠完全的消除回音，滿足智慧型行動裝置使用者的需求。經過模擬，在單端發話或者是雙端對講的情況都能夠有效地消除回音，跟其他架構相較之下，係數估測誤差(Misalignment)能夠減少20dB，此外，回音耗損增益(Echo Return Loss Enhancement , 簡稱ERLE)也提升了20dB，使多數的智慧型行動裝置使用者在進行通話不會受到回音的影響，而提升整個通話品質擁有較好的聽覺感受。

In this thesis, a dual microphone adaptive echo cancellation algorithm (AEC) is proposed. In the case of double talk, using an additional microphone is to provide added information about the signal which is the utilized to obtain an incoming signal estimate. This estimate signal is removed primary microphone’s near end speech to create the error signal which adapts the adaptive filter’s coefficients. The propose of using dual microphone is to overcome the problem remaining in using adaptive filters for echo cancellation. One is the biased estimate of adaptive filter’s coefficients because of the incoming speech contains near end speech components. The other is that adaptive filter’s coefficients may variate according to the change of incoming speech. In addition, the proposed in the single talk situation can completely eliminate the echo and meet the needs of smart handheld devices users. Finally, the results show that the proposed AEC is better than other AEC by 20 dB in misalignment and 20 dB extra added echo return loss enhancement (ERLE) is obtained. Therefore, users using the smart handheld devices will not be affected by the echo and enhance the quality of speech signal.

[1] eMaketer：台灣智慧型手機普及率達73.4%居全球之首. Available: https://read01.com/mJyKJO.html
[2] 各國智慧型手機普及率列表. Available: https://zh.wikipedia.org/wiki/%E5%90%84%E5%9C%8B%E6%99%BA%E6%85%A7%E5%9E%8B%E6%89%8B%E6%A9%9F%E6%99%AE%E5%8F%8A%E7%8E%87%E5%88%97%E8%A1%A8
[3] C. Hofmann, M. Guenther, C. Huemmer, and W. Kellermann, "Efficient nonlinear acoustic echo cancellation by partitioned-block Significance-Aware Hammerstein Group Models," in Signal Processing Conference (EUSIPCO), 2016 24th European, 2016, pp. 1783-1787: IEEE.
[4] J. H. Cho, D. R. Morgan, and J. Benesty, "An objective technique for evaluating doubletalk detectors in acoustic echo cancelers," IEEE Transactions on Speech and Audio Processing, vol. 7, no. 6, pp. 718-724, 1999.
[5] 國家通訊委員會105/09/26 新聞稿. Available: http://www.ncc.gov.tw/chinese/news_detail.aspx?site_content_sn=8&is_history=0&pages=1&sn_f=36281
[6] C. Breining et al., "Acoustic echo control. An application of very-high-order adaptive filters," IEEE Signal Processing Magazine, vol. 16, no. 4, pp. 42-69, 1999.
[7] P. Brady and G. Helder, "Echo suppressor design in telephone communications," The Bell System Technical Journal, vol. 42, no. 6, pp. 2893-2917, 1963.
[8] S. Weinstein, "Echo cancellation in the telephone network," IEEE Communications Society Magazine, vol. 15, no. 1, pp. 8-15, 1977.
[9] J. Benesty, T. Gänsler, D. R. Morgan, M. M. Sondhi, and S. L. Gay, Advances in network and acoustic echo cancellation. Springer, 2001.
[10] M. A. Iqbal and S. L. Grant, "Novel variable step size NLMS algorithms for echo cancellation," in Acoustics, Speech and Signal Processing, 2008. ICASSP 2008. IEEE International Conference on, 2008, pp. 241-244: IEEE.
[11] H. Ye and B.-X. Wu, "A new double-talk detection algorithm based on the orthogonality theorem," IEEE Transactions on Communications, vol. 39, no. 11, pp. 1542-1545, 1991.
[12] J. Benesty, D. R. Morgan, and J. H. Cho, "A new class of doubletalk detectors based on cross-correlation," IEEE Transactions on Speech and Audio Processing, vol. 8, no. 2, pp. 168-172, 2000.
[13] T.-A. Vu, H. Ding, and M. Bouchard, "A survey of double-talk detection schemes for echo cancellation applications," Canadian Acoustics, vol. 32, no. 3, pp. 144-145, 2004.
[14] S. S. Haykin, Adaptive filter theory. Pearson Education India, 2008.
[15] S. G. Sankaran and A. L. Beex, "Convergence behavior of affine projection algorithms," IEEE Transactions on Signal Processing, vol. 48, no. 4, pp. 1086-1096, 2000.
[16] J.-I. Nagumo and A. Noda, "A learning method for system identification," IEEE Transactions on Automatic Control, vol. 12, no. 3, pp. 282-287, 1967.
[17] D. Duttweiler, "A twelve-channel digital echo canceler," IEEE Transactions on Communications, vol. 26, no. 5, pp. 647-653, 1978.
[18] B. Porat, "Second-order equivalence of rectangular and exponential windows in least-squares estimation of Gaussian autoregressive processes," IEEE transactions on acoustics, speech, and signal processing, vol. 33, no. 5, pp. 1209-1212, 1985.
[19] B. Farhang-Boroujeny, Adaptive filters: theory and applications. John Wiley & Sons, 2013.
[20] G. Szwoch, A. Czyżewski, and M. Kulesza, "A low complexity double-talk detector based on the signal envelope," Signal Processing, vol. 88, no. 11, pp. 2856-2862, 2008.
[21] K. Pu, J. Zhang, and L. Min, "A signal decorrelation PNLMS algorithm for double-talk acoustic echo cancellation," Circuits, Systems, and Signal Processing, vol. 35, no. 2, pp. 669-684, 2016.
[22] M. Hamidia and A. Amrouche, "A new robust double-talk detector based on the Stockwell transform for acoustic echo cancellation," Digital Signal Processing, vol. 60, pp. 99-112, 2017.
[23] A. Carroll and G. Heiser, "An Analysis of Power Consumption in a Smartphone," in USENIX annual technical conference, 2010, vol. 14, pp. 21-21: Boston, MA.
[24] Q. Jingfan and O. Jingzheng, "A Novel Variable Step Size LMS Adaptive Filtering Algorithm Based on Sigmoid Function [J]," Journal of Data Acquisition & Processing, vol. 3, p. 002, 1997.
[25] P. C. Loizou, Speech enhancement: theory and practice. CRC press, 2013.
[26] E. Rothauser, "IEEE recommended practice for speech quality measurements," IEEE Trans. on Audio and Electroacoustics, vol. 17, pp. 225-246, 1969.
[27] 工研院文字轉語音Web服務. Available: http://tts.itri.org.tw/
[28] (2008). EBU Tech 3253 Users' Handbook for the EBU SQAM CD. Available: https://tech.ebu.ch/publications/tech3253