基於隱藏式半馬可夫模型(Hidden Semi-Markov Model)的文句轉語音系統(text-to-speech system)，是以統計模型描述語音的合成單元及其狀態時長，將輸入的文句表示成一序列的語音合成單元，然後轉換成語音輸出。改變語音合成單元的模型參數，就可以改變合成的聲音，因此可以利用模型調適方法，使合成的語音接近於目標語者的聲音特質、情緒特徵或說話韻律節奏，達到聲音轉換(voice conversion)的目的。
本論文更進一步利用目標語者語音的剩餘訊號(residual signal)，加入其語音產生模型的激發訊號中，使合成語音更接近目標語者聲音。論文中提出兩種剩餘訊號加入的方法，並對於合成的語音進行主觀評量與客觀評量。在主觀實驗中，發現其中一種剩餘訊號加入法會在聽覺上感覺到不連續聲音，而另一種方法則沒有不連續的狀況。在客觀評量中則是計算合成語音與目標語者語音的高斯混合模型，量測各高斯混合模型之間的KL距離，看出兩種加入剩餘訊號的方法，都使得其合成語音更接近目標語者語音。

[1] M. Abe, S. Nakamura, K. Shikano, and H. Kuwabara. "Voice conversion through vector quantization. " J. Acoust. Soc. Jpn. (E), Vol. 11, No. 2, pp.71–76, 1990.
[2]T. Toda, A.W. Black, K. Tokuda. "Voice conversion based on maximum likelihood estimation of spectral parameter trajectory. " IEEE Transactions on Audio, Speech and Language Processing, Vol. 15, No. 8, pp. 2222-2235, Nov. 2007.
[3] M. Charlier, Y. Ohtani, T. Toda, A. Moinet, T. Dutoit. "Cross-language voice conversion based on eigenvoices. " Proc. INTERSPEECH, pp. 1635-1638, Brighton, UK, Sep. 2009.
[4] A. Kain and M.W. Macon. "Spectral voice conversion for text-to-speech synthesis. " Proc. ICASSP, pp. 285–288, Seattle, USA, May 1998.
[5] A. Hunt and A.W. Black, "Unit selection in a concatenative speech synthesis system using a large speech database." in Proc. ICASSP, 1996, pp. 373-376.
[6] H. Zen, K. Tokuda, T. Masuko, T. Kobayashi, and T. Kitamura, "Hidden semi-Markov model based speech synthesis. "¸in Proc. of ICSLP, 2004, pp. 1185–1180.
[7] K. Tokuda ,T. Yoshimura, T. Masuko, T. Kobayashi, and T. Kitamura. "Speech parameter generation algorigthms for HMM-based speech synthesis. " In Proc. ICASSP 2000, pages 1315–1318, June 2000.
[8] T. Toda and K. Tokuda, “A speech parameter generation algorithm considering global variance for HMM-based speech synthesis,” IEICE Trans. Inf. & Syst., vol. E90-D, no. 5, pp. 816–824, May 2007
[9] S. Imai, “Cepstral analysis synthesis on the mel frequency scale,”Proc. of ICASSP, pp.93–96, 1983.
[10]J.Yamagishil,H.Zen,T Toda and K. Tokuda ,“Speaker-Independent HMM-based Speech Synthesis System— HTS-2007 System for the Blizzard Challenge 2007.”Blizzard 2007,2007
[11] T. Kobayashi and S. Imai, “Spectral analysis using generalized cepstrum,”IEEE Trans. Acoust., Speech, Signal Processing, vol. ASSP-32,
pp.1087–1089, Oct. 1984.
[12] K. Tokuda, T. Kobayashi, T. Masuko and S. Imai,” Mel-generalized cepstral analysis— a unified approach to speech spectral estimation.” Proc.ICASSP,pp.1043-1064,1994
[13]G. Muhammad“Extended Average Magnitude Difference Function Based Pitch Detection“ IAJIT Vol. 8, No. 2, April 2011 pp.197-203

[14] K. Tokuda, T. Masuko, N. Miyazaki, and T. Kobayashi. Hidden Markov
models based on multi-space probability distribution for pitch pattern
modeling. In Proc. ICASSP-99, pages 229–232, March 1999.
[15] T. Yoshimuray, K.Tokuday, T. Masukoyy, T. Kobayashiyy and T. Kitamuray.” Simultaneous modeling of spectrum, pitch and duration in hmm-based speech synthesis” Proc. of Eurospeech 1999. Budapest, Hungary.
[16] H. Zen, T. Masuko, T. Yoshimura, K. Tokuda, T. Kobayashi, and
T. Kitamura, “State duration modeling for HMM-based speech synthesis,” IEICE Trans. on Inf. & Syst., vol. E90-D, no. 3, pp.692–693, 2007.
[17] K. Shinoda and T. Watanabe. MDL-based context-dependent subword
modeling for speech recognition. J. Acoust. Soc. Japan (E), 21:79–86,March 2000.
[18] L.F. Uebel, P.C. Woodland, “An Investigation into Vocal Tract Length Normalisation”, Proc. Eurospeech, Vol. 6, pp. 2527- 2530, Budapest, Hungary, Sep. 1999.
[19] C.J. Leggetter & P.C. Woodland.”Maximum Likelihood Linear Regression for Speaker Adaptation of Continuous Density Hidden Markov Models.” Computer Speech & Language, Vol. 9, pp. 171-185.
[20] O. Shiohan, T. Myrvoll, and C. Lee, “Structural maximum a posteriori
linear regression for fast HMM adaptation,” Computer Speech & Language,
vol. 16, no. 3, pp. 5–24, 2002.
[21] J. Yamagishi, T. Kobayashi, Y.Nakano,K.Ogata and J.Isogai “Analysis of Speaker Adaptation Algorithms for HMM-Based Speech Synthesis and a Constrained SMAPLR Adaptation Algorithm .“, IEEE.Speech,Audio,Lang.Process., Vol.17 JANUARY 2009
[22] O. Shiohan, T. Myrvoll, and C. Lee, “Structural maximum a posteriori linear regression for fast HMM adaptation,” Computer Speech & Language, vol. 16, no. 3, pp. 5–24, 2002.

[23] Jacob Goldberger, Shiri Gordon, and Hayit Greenspan, “An efficient image similarity measure based on approximations of KL-divergence between two gaussian mixtures,” in Proc. of ICCV 2003, Nice, October 2003, vol. 1, pp. 487–493.
[24] John R. Hershey and Peder A. Olsen “Approximating the kullback leibler divergence between gaussian mixture models”in ICASSP 2007,vol 4 pp. 317-320
[25] Speech Signal Processing Toolkit (SPTK)
http://sp-tk.sourceforge.net/
[26] The Hidden Markov Model Toolkit (HTK)
http://htk.eng.cam.ac.uk/
[27] HMM-based Speech Synthesis System (HTS)
http://hts.sp.nitech.ac.jp/
[28] HTS Engine
http://hts-engine.sourceforge.net/
[29]wavesurfer
http://www.speech.kth.se/wavesurfer/
[30]中研院斷詞系統
http://ckipsvr.iis.sinica.edu.tw/
[31] 王小川,語音訊號處理, 全華圖書公司 2007