本論文研究製作迴音消除器與噪音刪減器軟體，然後放入WinRTP網路電話軟體裡，以提升網路電話的通話品質。在噪音刪減方面，先改進亂度偵測法來作語音活動偵測及估計雜訊頻譜，再研究以非線性頻譜刪減法和改良式頻譜刪減法，進行兩段式的頻譜刪減動作，來削減雜訊訊號。在迴音消除方面，採用正規化最小均方演算法製作一適應性濾波器，用以濾除迴音訊號，並且藉由分析電腦處理網路封包所導入之延遲時間，來降低適應性濾波器的計算量與收斂速度。依據實驗量測到的訊噪比與主觀聽測的評估分數，顯示我們的雜訊刪減器和迴音消除器的確可讓噪音訊號與迴音訊號都大幅的衰減，而使通話語音之品質獲得明顯的提升。

In this paper, we study the production of acoustic echo cancellation and noise reduction software, and then into WinRTP Internet phone software, to improve the quality of internet phone calls. In the area of noise reduction, first to improve the degree of entropy detection method for voice activity detection, then noise spectrum is estimated. We combine non-linear spectrum subtraction with improved spectrum subtraction, two-stage action of the spectrum subtraction, to reduce the noise. In the echo cancellation, we use normalized least mean square adaptive algorithm to filter out echo signals, and by analysis of the delay to reduce computation load and convergence speed. Based on the experimental measurement to the hearing and subjective assessment of test scores, we show that the noise subtraction and echo cancellation device significantly attenuate the noise and echo signal, and therefore enhance the quality of voice.

[1]Regine Le Bouquin Jeannes, Pascal Scalart, Gerard Faucon, Christophe Beaugeant, “Combined Noise and Echo Reduction in Hands-Free System:A Survey”, Speech and Audio processing, Vol. 9, No. 8, pp. 808-820, 2001.
[2]M. M. Sondhi, “An adaptive echo canceler”, Bell System Techn.J, Vol XLVI, No. 3, pp. 497-510, 1967.
[3]C. Gritton, D. Lin, “Echo cancellation algorithms”, IEEE ASSP Magazine, Vol. 1, pp. 30-38, 1984.
[4]C. Breining, P. Dreiscitel, E. Hansler, A. Mader, et al, “Acoustic Echo Control.”, IEEE Signal Processing Magazine, Vol. 16, pp. 42-69, 1999.
[5]Per Ahgren, Andreas Jakobsson, “A Study of Doubletalk Detection Performance in the Presence of Acoustic Echo Path Changes.”, IEEE trans. on Consumer Electronics, Vol. 52,No. 2, pp. 515-522, 2006.
[6]ITU, The silence compression scheme for G.729 optimized for terminals conforming to Recommendation V. 70 , ITU-T-Recommendations G.729 Annex B.
[7]賴辰瑋，強健性語音辨識之研究:語音前端端點偵測與語音強化法，國立暨南國際大學電機工程研究所，2004。
[8]Philippe Renevery, Andrzej Drygajlo, “Entropy Based Voice Activity Detection in Very Noise Conditions.”, European Conference on Speech Communication and Technology, Eurospeech, 2001.
[9]Shannon, C. E., “A mathematical theory of communication”, Bell System Technical Journal, Vol. 27, pp. 379-423, 623-656, July, Oct, 1984.
[10]Abdallah I.,Montresor S., and Baudry M., ”Speech signal detection in noisy environment using a local entropic criterion”, Eurospeech, Rhodes, Greece, Sep, 1997.
[11]Boll, S. F, ”Suppression of Acoustic Noise in Speech Using Spectral Subtraction.”, IEEE Trans. On ASSP, Vol. 27, NO. 2, pp. 113-120, 1979.
[12]R. M. Udera, S. Ciochina, ”Speech Enhancement using Spectral Over-subtraction and Residual Noise Reduction”, Signals, Circuits and Systems, IEEE, pp. 165-168, 2003.
[13]Norihide Kitaoka and Seiichi Nakagawa, “Evaluation of Spectral Subtraction with Smoothing of Time Direction on the Aurora 2 Task.”, ICSLLP, 2002.
[14]王小川，語音訊號處理，全華出版社。
[15]郭振甫，應用數位訊號處理器於聲學迴音消除，國立成功大學航空太空工程研究所，2002。
[16]R. Storn, “Echo Cancellation Techniques for Multimedia Applications – a Survey”, International Computer Science Institutek, Berkeley, Nov, 1996.
[17]Youhong Lu, Wenshun Tian, “Enhancing Echo Cancellation via Estimation of Delay.”, IEEE Trans. on signal processing, Vol. 53, No. 11, Novermver, 2005.
[18]鄭聖穎，聯合聲學迴音消除與雙向通話偵測演算法及其於定點式DSP之實現，國立雲林科技大學電子工程研究所，2005。
[19]Regine Le Bouquin Jeannes, Pascal Scalart,Gerard Faucon, Christophe Beaugeant, “Combined Noise and Echo Reduction in Hands-Free Systems: A Survey.”, IEEE Transactions on speech and audio processing, Vol. 9, No. 8, November, 2001.
[20]洪維廷、陳弘啟、陳軍廷，”麥克風陣列技術應用於語音加強及語音辨識之研究”，全國電信研討會，2008。
[21]Giorgio Zoia, Alain Sturzenegger, Olivier Hochreutiner, “Audio Quality and Acoustic Echo Issues for VOIP on Portable Devices.“, Porable information devices IEEE internationl conference, 2007.
[22]WinRTP, Vovida, http://www.vovida.org/.