A Higher Intelligibility Speech-Enhancement Algorithm

Abstract:

A higher intelligibility speech-enhancement algorithm based on subspace is proposed. The majority existing speech-enhancement algorithms cannot effectively improve enhanced speech intelligibility. One important reason is that they only use Minimum Mean Square Error (MMSE) to constrain speech distortion but ignore that speech distortion region differences have a significant effect on intelligibility. A priori Signal Noise Ratio (SNR) and gain matrix were used to determine the distortion region. Then the gain matrix was modified to constrain the magnitude spectrum of the amplification distortion in excess of 6.02 dB which damages intelligibility much. Both objective evaluation and subjective audition show that the proposed algorithm does improve the enhanced speech intelligibility.

• Periodical: Applied Mechanics and Materials (Volumes 321-324)
• Pages: 1075-1079
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.321-324.1075
• Online since: June 2013
• Authors: Peng Liu, Jian Fen Ma
• Keywords: Gain Matrix, Objective Evaluation, Speech Intelligibility, Subjective Audition

References

[1] Philipos C. Loizou: Speech Enhancement: Theory and Practice [M]. Boca Raton: Florida: CRC Press LLC(2007).

[2] Philipos C. Loizou, Gibak Kim: Reasons why current speech-enhancement algorithms do not improve speech intelligibility and suggested solutions [J]. IEEE transactions on audio, speech, and language processing, Vol. 19, No. 1 (2011), pp.47-56

DOI: 10.1109/tasl.2010.2045180

[3] Hu, Y. , Loizou, P.: A generalized subspace approach for enhancing speech corrupted by colored noise[J].IEEE Trans. on Speech and Audio Processing,Vol.11, No.4(2003), pp.334-341

DOI: 10.1109/tsa.2003.814458

[4] Scalart P, Vieira-Filho J: Speech enhancement based on a priori signal to noise estimation[C].//Proc 21st IEEE Int Conf Acoust Speech Signal Processing, Atlanta, GA, Vol.2, No.2(1996), pp.629-632.

DOI: 10.1109/icassp.1996.543199

[5] Lu Y., Loizou P. :Speech enhancement by combining statistical estimators of speech and noise[J] in Proceedings of IEEE International Conference on Acoustics, Speech, Signal Processing (2010), p.4754–4757

DOI: 10.1109/icassp.2010.5495156

[6] ZHANG Hua,WANG Shuo,CHEN Jing et al:The Manda-rin Speech Test Materials (MSTMs): Development and application[J]. Chinese Scientific Journal of Hearing and Speech Rehabilitation, Vol. 6(2008), pp.16-17. ( In Chinese)

[7] Ma, J., Hu, Y., Loizou, P.: Objective measures for pre-dicting speech intelligibility in noisy conditions based on new band-importance functions[J]. J. Acoust. Soc. Am., Vol. 125, No. 5(2009), pp.3387-3405.

DOI: 10.1121/1.3097493

[8] ITU-T P.862(2001), Perceptual evaluation of speech quality (PESQ): An objective method for end-to-end speech quality assessment of narrowband telephone networks and speech codecs [S]

DOI: 10.1109/icassp.2001.941023