Speech enhancement for non-stationary noise environments

Cohen, Israel; Berdugo, Baruch

doi:10.1016/s0165-1684(01)00128-1

Cited by 529 publications

(374 citation statements)

References 23 publications

Supporting

Mentioning

362

Contrasting

Unclassified

Order By: Relevance

“…In practice, the noise signal is unknown, and the noise spectral variance can be estimated by using the Minima Controlled Recursive Averaging approach [17]. Furthermore, the speech presence probability is estimated from the noisy spectral measurements [6]. Table 2 shows the results of the LSD obtained by using the different algorithms for various SNR levels.…”

Section: Resultsmentioning

confidence: 99%

“…In the other time-frequency bins,p tk is set to zero. Furthermore, we assume knowledge of the noise variance s 2 tk , which in practice can be estimated by using the Minima Controlled Recursive Averaging approach [6,17]. Our objectives in this work are as follows:…”

Section: Article In Pressmentioning

confidence: 99%

“…It enables to derive useful minimum meansquared error (MMSE) estimators for the shortterm spectral amplitude (STSA), as well as the logspectral amplitude (LSA) [2,3], and it underlies the design of many speech enhancement algorithms, e.g. [4][5][6][7][8]. Martin [9] considered a Gamma speech model, under which the real and imaginary parts of the STFT coefficients are modeled as independent and identically distributed (iid) Gamma random variables.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Speech spectral modeling and enhancement based on autoregressive conditional heteroscedasticity models

Cohen

2006

Signal Processing

Self Cite

View full text Add to dashboard Cite

In this paper, we develop and evaluate speech enhancement algorithms, which are based on supergaussian generalized autoregressive conditional heteroscedasticity (GARCH) models in the short-time Fourier transform (STFT) domain. We consider three different statistical models, two fidelity criteria, and two approaches for the estimation of the variances of the STFT coefficients. The statistical model is either Gaussian, Gamma or Laplacian; the fidelity criteria include minimum mean-squared error (MMSE) of the STFT coefficients and MMSE of the log-spectral amplitude (LSA); the spectral variance is estimated based on either the proposed GARCH models or the decision-directed method of Ephraim and Malah. We show that estimating the variance by the GARCH modeling method yields lower log-spectral distortion and higher perceptual evaluation of speech quality scores (PESQ, ITU-T P.862) than by using the decisiondirected method, whether the presumed statistical model is Gaussian, Gamma or Laplacian, and whether the fidelity criterion is MMSE of the STFT coefficients or MMSE of the LSA. furthermore while a gaussian model is inferior to the supergaussian models when USING the decision-directed method, the Gaussian model is superior when using the garch modeling method. r 2005 Published by Elsevier B.V.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Article In Pressmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Speech spectral modeling and enhancement based on autoregressive conditional heteroscedasticity models

Cohen

2006

Signal Processing

Self Cite

View full text Add to dashboard Cite

show abstract

“…이러한 음성 향상에서는 목표가 되는 음성 과 그 외의 잡음을 각기 다른 가우시안 통계 모델로 만들고 매 시간 프레임 마다 음성 존재 검출(Voice Activity Detection, VAD) 방법 등을 결합해서 향상 과정을 수행 한다 [2,3] . 이러한 방법의 향상은 잡음이 시간에 흐름에 따라 천천히 변한다는 가정을 한다.…”

Section: ⅰ 서 론unclassified

A NMF-Based Speech Enhancement Method Using a Prior Time Varying Information and Gain Function

Kwon¹,

Jin²,

Bae³

et al. 2013

The Journal of Korean Institute of Communications and Informati

View full text Add to dashboard Cite

This paper presents a speech enhancement method using non-negative matrix factorization. In training phase, we can obtain each basis matrix from speech and specific noise database. After training phase, the noisy signal is separated from the speech and noise estimate using basis matrix in enhancement phase. In order to improve the performance, we model the change of encoding matrix from training phase to enhancement phase using independent Gaussian distribution models, and then use the constraint of the objective function almost same as that of the above Gaussian models. Also, we perform a smoothing operation to the encoding matrix by taking into account previous value. Last, we apply the Log-Spectral Amplitude type algorithm as gain function.

show abstract

“…The preprocessor is realized as a speech enhancement module because raw speech data is the only format which most existing speech recognition systems accept, despite efforts to standardize the feature for speech recognition [11]. As the starting point, we have chosen the optimally modified log spectral amplitude (OM-LSA) speech estimator, combined with the minima-controlled recursive averaging (MCRA) noise estimator developed by Cohen and Berdugo [12]. It was confirmed by objective and subjective tests [13] that the OM-LSA speech estimator can achieve satisfactory suppression of noise with only a slight distortion of the signal.…”

Section: Introductionmentioning

confidence: 99%

Noise suppression method for preprocessor of time-lag speech recognition system based on bidirectional optimally modified log spectral amplitude estimation

Obuchi

Takeda

Togami

2013

Acoust. Sci. & Tech.

View full text Add to dashboard Cite

In this paper, we propose a new noise suppression method, that is best used as a preprocessor for time-lag speech recognition. Assuming that a time lag of a few seconds is acceptable in various speech recognition applications, the proposed method is realized as a combination of forward and backward estimation flows over time. Each estimation flow is based on the optimally modified log spectral amplitude (OM-LSA) speech estimator, but a look-ahead estimation mechanism is additionally equipped to make the estimation more robust. Evaluation experiments using various databases confirm that the speech recognition accuracy can be greatly improved by adding the proposed method to the existing system.

show abstract

Speech enhancement for non-stationary noise environments

Cited by 529 publications

References 23 publications

Speech spectral modeling and enhancement based on autoregressive conditional heteroscedasticity models

Speech spectral modeling and enhancement based on autoregressive conditional heteroscedasticity models

A NMF-Based Speech Enhancement Method Using a Prior Time Varying Information and Gain Function

Noise suppression method for preprocessor of time-lag speech recognition system based on bidirectional optimally modified log spectral amplitude estimation

Contact Info

Product

Resources

About