在本篇論文中，一個基於人耳聽覺的特徵演算法將被提出用於強健語音的辨識系統上。在此篇研究中，語音訊號將會經由一個新的特徵演算法稱做人耳基底膜頻帶倒頻譜係數(Basilar-membrane Frequency-band Cepstral Coefficient, BFCC)被特徵化，與現在普遍使用的梅爾倒頻譜係數(Mel Frequency Cepstral Coefficient, MFCC)方法相比，MFCC使用複利葉轉換來產生頻譜而BFCC則是使用基於人耳基底膜的濾波器(gammachirp)的小波轉換來產生頻譜，因為梅爾三角濾波器與人耳基底膜濾波器的不同以及複利葉轉換的特性與小波轉換特性的不同使得BFCC產生的頻譜可以更準確的模仿人耳聽覺的特性以及改善雜訊的干擾。此外，本篇論文使用HTK工具在訓練以及測試時產生隱藏式馬可夫模型(Hidden Markov Model, HMM)。本篇論文使用AURORA 2.0做為訓練以及測試時的資料庫，測試使用AURORA 2.0裡的testA做測試資料，雜訊分別有列車、人聲、車子、展廳，辨識結果顯示出所提出的BFCC方法比MFCC方法在雜訊比範圍-5dB到20dB裡，平均四種雜訊的語音辨識率改善了13%，並與其他人耳聽覺特徵(Gammatone Wavelet Cepstral Coefficient, GWCC)、(Gammatone Frequency Cepstral Coefficient, GFCC)做比較，平均語音辨識率與GWCC相比約改善17%，與GFCC辨識率相比約改善0.5%。

An auditory-based feature extraction algorithm is proposed for enhancing the robustness of automatic speech recognition. In the proposed approach, the speech signal is characterized using a new feature referred to as the Basilar-membrane Frequency-band Cepstral Coefficient (BFCC). In contrast to the conventional Mel-Frequency Cepstral Coefficient (MFCC) method based on a Fourier spectrogram, the proposed BFCC method uses an auditory spectrogram based on a gammachirp wavelet transform in order to more accurately mimic the auditory response of the human ear and improve the noise immunity. In addition, a Hidden Markov Model (HMM) is used for both training and testing purposes. The evaluation results obtained using the AURORA 2 noisy speech database show that compared to the MFCC method, Gammatone Wavelet Cepstral Coefficient (GWCC), and Gammatone Frequency Cepstral Coefficient (GFCC), the proposed scheme improves the speech recognition rate by 13%, 17%, and 0.5% on average given speech samples with Siganl-to-Noise Ratios (SNRs) ranging from -5 to 20 dB, respectively.

[1]S. Boll, “Suppression of acoustic noise in speech using spectral subtraction,” Acoustics, Speech and Signal Processing, IEEE Transactions , vol. 27, no. 2, pp. 113–120, Apr. 1979.
[2]J. S. Goldstein, S. I. Reed, L. L. Scharf, “A multistage representation of the Wiener filter based on orthogonal projections,” Information Theory, IEEE Transactions , vol. 44, no. 7, pp. 2943–2959, Nov. 1998.
[3]W. Han, C.-F. Chan, C.-S. Choy, K.-P. Pun, “An efficient MFCC extraction method in speech recognition,” in Pro. IEEE International Symposium on Circuits and Systems, 2006, May 21–24, pp. 4.
[4]R. Patterson and I. N. Smith, “An efficient auditory filterbank based on the gammatone function,” Speech-Group meeting of the Institute of Acoustics on Auditory Modelling, vol. 54, Apr. 1987.
[5]A. Adiga, M. Magimai, C. S. Seelamantula, “Gammatone wavelet Cepstral Coefficients for robust speech recognition,” IEEE Region 10 Conference on TENCON, 2013, Oct. 22–25, pp. 1–4, 22–25.
[6]Y. Shao; S. Srinivasan, D. Wang, “Incorporating Auditory Feature Uncertainties in Robust Speaker Identification,” Acoustics, Speech and Signal Processing, 2007. ICASSP 2007. IEEE International Conference, 2007, Apr. 15–20, vol. 4, pp. 277–280.
[7]X. Yang, K. Wang, S. A. Shamma, “Auditory representations of acoustic signals,” Information Theory, IEEE Transactions, vol. 38, no. 2, pp. 824–839, Mar. 1992.
[8]L. Solbach, R. Wohrmann, and J. Kliewer, “The complex-valued continuous wavelet transform as a preprocessor for auditory scene analysis,” in H. Okuno and D. Rosenthal (editors): Readings in Computational Auditory Scene Analysis, pp. 273–292, Erlbaum Publishers, 1998.
[9]Q. Li, “An auditory-based transfrom for audio signal processing,” Applications of Signal Processing to Audio and Acoustics, 2009. WASPAA '09. IEEE Workshop, 2009 Oct. 18-21, pp. 181–184.
[10]Q. Li, Y. Huang, “An auditory-based feature extraction algorithm for robust speaker identification under mismatched conditions,” IEEE Transactions on Audio, Speech, and Language Processing, vol. 19, no. 6, pp. 1791–1801, Aug. 2011.
[11]A. Mertins, J. Rademacher, “Vocal tract length invariant features for automatic speech recognition,” IEEE Workshop on Automatic Speech Recognition and Understanding, 2005, Nov. 27–27, pp. 308–312.
[12]A. Venkitaraman, A. Adiga, and C. S. Seelamantula, “Auditory-motivated gammatone wavelet transform,” Signal Processing, vol. 94, pp. 608–619, Jan. 2014.
[13]B. Hanson, T. Applebaum, “Robust speaker-independent word recognition using static, dynamic and acceleration features: experiments with Lombard and noisy speech,” International conference on acoustics, speech and signal processing, 1990, Apr. 3–6, pp. 857–860, vol. 2.
[14]D. Schofield, “Visualisations of speech based on a model of the peripheral auditory system,” NASA STI/Recon Technical Report N, vol.86, pp. 17593, 1985.
[15]R. D. Patterson, “Auditory filter shapes derived with noise stimuli,” The Journal of the Acoustical Society of America, vol.59, no.3, pp. 640–654, Mar. 1976.
[16]R. D. Patterson et al. “An efficient auditory filterbank based on the gammatone function,” a meeting of the IOC Speech Group on Auditory Modelling at RSRE, vol. 2, no. 7, Dec. 1987.
[17]T. Irino and R. D. Patterson, “A time-domain, level-dependent auditory filter: The gammachirp,” The Journal of the Acoustical Society of America, vol.101, no.1, pp. 412–419, Jan. 1997.
[18]R. A. Lutfi and R. D. Patterson, “On the growth of masking asymmetry with stimulus intensity,” The Journal of the Acoustical Society of America, vol.76, no.3, pp. 739–745, Sep. 1984.
[19]B. C. J. Moore, R. W. Peters, and B. R. Glasberg, “Auditory filter shapes at low center frequencies,” The Journal of the Acoustical Society of America, vol.88, no.1, pp. 132–140, July 1990.
[20]S. Rosen and R. J. Baker, “Characterising auditory filter nonlinearity,” Hearing research, vol.73, no.2, pp. 231–243, Mar. 1994.
[21]T. Irino and R. D. Patterson, “A compressive gammachirp auditory filter for both physiological and psychophysical data,” The Journal of the Acoustical Society of America, vol.109, no.5, pp. 2008–2022, May 2001.
[22]L. H. Carney, M. J. McDuffy, and I. Shekhter, “Frequency glides in the impulse responses of auditory-nerve fibers,” The Journal of the Acoustical Society of America, vol.105, no.4, pp. 2384–2391, Apr. 1999.
[23]L. Sun, M. Shen, F. H. Y. Chan, and P. J. Beadle, “Instantaneous Frequency Estimate of Nonstationary Phonocardiograph Signals Using Hilbert Spectrum,” Engineering in Medicine and Biology Society, 2005. IEEE-EMBS 2005. 27th Annual International Conference, 2006, Jan. 17–18, pp. 7285–7288.
[24]Y. Litvin, I. Cohen, and D. Chazan, “Separation of speech and music sources from a single-channel mixture using discrete energy separation algorithm,” in International Workshop on Acoustic Echo and Noise Control, IWAENC, 2010.
[25]E. Bedrosian, “A product theorem for Hilbert transforms,” Proc. IEEE, 1963, May, pp. 868–869, vol. 51.
[26]I. Daubechies, “The wavelet transform, time-frequency localization and signal analysis,” Information Theory, IEEE Transactions, vol. 36, no. 5, pp. 961–1005, Sep 1990.