Filtering of colored noise for speech enhancement and coding

Scalar and vector Kalman filters are implemented for filtering speech contaminated by additive white noise or colored noise, and an iterative signal and parameter estimator which can be used for both noise types is presented. Particular emphasis is placed on the removal of colored noise, such as helicopter noise, by using state-of-the-art colored-noise-assumption Kalman filters. The results indicate that the colored noise Kalman filters provide a significant gain in signal-to-noise ratio (SNR), a visible improvement in the sound spectrogram, and an audible improvement in output speech quality, none of which are available with white-noise-assumption Kalman and Wiener filters. When the filter is used as a prefilter for linear predictive coding, the coded output speech quality and intelligibility are enhanced in comparison to direct coding of the noisy speech. >

[1]  M. Pagano.  Estimation of Models of Autoregressive Signal Plus White Noise , 1974 .

[2]  Amir Dembo.  Signal reconstruction from noisy partial information of its transform , 1989, IEEE Trans. Acoust. Speech Signal Process..

[3]  Ehud Weinstein,et al.  Parameter estimation of superimposed signals using the EM algorithm , 1988, IEEE Trans. Acoust. Speech Signal Process..

[4]  Benjamin Friedlander,et al.  A spectral matching technique for ARMA parameter estimation , 1983, ICASSP.

[5]  G. W. Hughes,et al.  Minimum Prediction Residual Principle Applied to Speech Recognition , 1975 .

[6]  Kuldip K. Paliwal.  A noise-compensated long correlation matching method for AR spectral estimation of noisy signals , 1986, ICASSP '86. IEEE International Conference on Acoustics, Speech, and Signal Processing.

[7]  K. Kroschel.  A comparison of quantizers optimized for corrupted and uncorrupted input signals , 1987 .

[8]  Hans S. Witsenhausen,et al.  Indirect rate distortion problems , 1980, IEEE Trans. Inf. Theory.

[9]  Jack K. Wolf,et al.  Transmission of noisy information to a noisy receiver with minimum distortion , 1970, IEEE Trans. Inf. Theory.

[10]  B.-H. Juang,et al.  On using the Itakura-Saito measures for speech coder performance evaluation , 1984, AT&T Bell Laboratories Technical Journal.

[11]  D. Rubin,et al.  Maximum likelihood from incomplete data via the EM - algorithm plus discussions on the paper , 1977 .

[12]  Thomas R. Fischer,et al.  Estimation and vector quantization of noisy speech , 1988, ICASSP-88., International Conference on Acoustics, Speech, and Signal Processing.

[13]  Kuldip K. Paliwal,et al.  A speech enhancement method based on Kalman filtering , 1987, ICASSP '87. IEEE International Conference on Acoustics, Speech, and Signal Processing.

[14]  Benjamin Friedlander,et al.  A spectral matching technique for ARMA parameter estimation , 1984 .

[15]  Robert M. Gray,et al.  An Algorithm for Vector Quantizer Design , 1980, IEEE Trans. Commun..

[16]  C. R. Rao,et al.  Linear Statistical Inference and its Applications , 1968 .

[17]  Joël Le Roux,et al.  Comparison of some algorithms for identifying autoregressive signals in the presence of observation noise , 1982, ICASSP.

[18]  New York Dover,et al.  ON THE CONVERGENCE PROPERTIES OF THE EM ALGORITHM , 1983 .

[19]  Charles F. Teacher,et al.  Performance of LPC vocoders in a noisy environment , 1979, ICASSP.

[20]  Nuggehally Sampath Jayant,et al.  LPC analysis/Synthesis from speech inputs containing quantizing noise or additive white noise , 1976 .

[21]  Ender Ayanoglu.  Optimal quantization of noisy sources , 1988, ICASSP-88., International Conference on Acoustics, Speech, and Signal Processing.

[22]  Jerry D. Gibson,et al.  Alphabet-constrained data compression , 1982, IEEE Trans. Inf. Theory.

[23]  B. Anderson,et al.  Optimal Filtering , 1979, IEEE Transactions on Systems, Man, and Cybernetics.

[24]  Andrew P. Sage,et al.  Estimation theory with applications to communications and control , 1979 .

[25]  John H. L. Hansen,et al.  Constrained iterative speech enhancement with application to speech recognition , 1991, IEEE Trans. Signal Process..

[26]  Robert M. Gray,et al.  A unified approach for encoding clean and noisy sources by means of waveform and autoregressive model vector quantization , 1988, IEEE Trans. Inf. Theory.

[27]  Craig K. Rushforth,et al.  Estimating the parameters of a noisy all-pole process using pole-zero modeling , 1979, ICASSP.

[28]  Alan V. Oppenheim,et al.  All-pole modeling of degraded speech , 1978 .

[29]  Bruce R. Musicus,et al.  Iterative algorithms for optimal signal reconstruction and parameter identification given noisy and incomplete data , 1983, ICASSP.

[30]  P. V. de Souza,et al.  LPC distance measures and statistical tests with particular reference to the likelihood ratio , 1982 .