Enhanced Speech Features by Single-Channel Joint Compensation of Noise and Reverberation

Wölfel, Matthias

doi:10.1109/tasl.2008.2009161

Cited by 44 publications

(37 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The convolutive nature of reverberation induces a long-term correlation between a current observation and past observations of reverberant speech. This longterm correlation has been exploited to mitigate the effect of reverberation directly on the speech signal (i.e., speech [9][10][11][12] or feature [13,14] dereverberation) or on the acoustic model used for recognition [15,16]. The REVERB challenge [17] was organized to evaluate recent progress in the field of reverberant speech enhancement (SE) and recognition.…”

Section: Y(t) = H(t) * S(t) + N(t)mentioning

confidence: 99%

Strategies for distant speech recognitionin reverberant environments

Delcroix

Yoshioka

Ogawa

et al. 2015

EURASIP J. Adv. Signal Process.

View full text Add to dashboard Cite

Reverberation and noise are known to severely affect the automatic speech recognition (ASR) performance of speech recorded by distant microphones. Therefore, we must deal with reverberation if we are to realize high-performance hands-free speech recognition. In this paper, we review a recognition system that we developed at our laboratory to deal with reverberant speech. The system consists of a speech enhancement (SE) front-end that employs long-term linear prediction-based dereverberation followed by noise reduction. We combine our SE front-end with an ASR back-end that uses neural networks for acoustic and language modeling. The proposed system achieved top scores on the ASR task of the REVERB challenge. This paper describes the different technologies used in our system and presents detailed experimental results that justify our implementation choices and may provide hints for designing distant ASR systems.

show abstract

Section: Y(t) = H(t) * S(t) + N(t)mentioning

confidence: 99%

Strategies for distant speech recognitionin reverberant environments

Delcroix

Yoshioka

Ogawa

et al. 2015

EURASIP J. Adv. Signal Process.

View full text Add to dashboard Cite

show abstract

“…Using interacting sub-models, the predictive PDF may be approximated by the following weighted sum of submodel specific predictive PDFs: (9) Here, indicates the index of the active submodel at time instant . Due to the definition of our state vector in (5), the sub-model specific PDF is completely determined by the PDF , which is approximated by (10) (11) In (10), denotes a GAUSSIAN PDF with mean vector and covariance matrix . For , the respective GAUSSIAN PDF for sub-model is characterized by the state transition matrix , the bias compensation vector and the linear prediction error covariance matrix .…”

Section: A a Priori Speech Modelmentioning

confidence: 99%

“…In [11] the joint compensation of reverberation and noise was considered in the feature domain, also within a BAYESIAN framework, however quite differently. It assumed reverberation being an additive distortion in the mel power spectral domain, employed a different a priori model and the approximate inference was realized by a particle filter.…”

Section: Introductionmentioning

confidence: 99%

Bayesian Feature Enhancement for Reverberation and Noise Robust Speech Recognition

Leutnant

Krueger

Haeb‐Umbach

2013

IEEE Trans. Audio Speech Lang. Process.

View full text Add to dashboard Cite

Abstract-In this contribution we extend a previously proposed BAYESIAN approach for the enhancement of reverberant logarithmic mel power spectral coefficients for robust automatic speech recognition to the additional compensation of background noise. A recently proposed observation model is employed whose time-variant observation error statistics are obtained as a side product of the inference of the a posteriori probability density function of the clean speech feature vectors. Further a reduction of the computational effort and the memory requirements are achieved by using a recursive formulation of the observation model. The performance of the proposed algorithms is first experimentally studied on a connected digits recognition task with artificially created noisy reverberant data. It is shown that the use of the time-variant observation error model leads to a significant error rate reduction at low signal-to-noise ratios compared to a time-invariant model. Further experiments were conducted on a 5000 word task recorded in a reverberant and noisy environment. A significant word error rate reduction was obtained demonstrating the effectiveness of the approach on real-world data.Index Terms-Robust automatic speech recognition, modelbased BAYESIAN feature enhancement, observation model for reverberant and noisy speech, recursive observation model.

show abstract

“…The method first estimates late reverberations using long-term MSLP, and then suppresses these with subsequent spectral subtraction. Wolfel proposed a joint compensation of noise and reverberation by integrating an estimate of the reverberation energy derived by an auxiliary model based on MSLP, into a framework, which so far, tracks and removes nonstationary additive distortion by particle filters in a low-dimension logarithmic power frequency domain [19].…”

Section: Introductionmentioning

confidence: 99%

Environment-dependent denoising autoencoder for distant-talking speech recognition

Ueda

Wang

Kai

et al. 2015

EURASIP J. Adv. Signal Process.

View full text Add to dashboard Cite

In this paper, we propose an environment-dependent denoising autoencoder (DAE) and automatic environment identification based on a deep neural network (DNN) with blind reverberation estimation for robust distant-talking speech recognition. Recently, DAEs have been shown to be effective in many noise reduction and reverberation suppression applications because higher-level representations and increased flexibility of the feature mapping function can be learned. However, a DAE is not adequate in mismatched training and test environments. In a conventional DAE, parameters are trained using pairs of reverberant speech and clean speech under various acoustic conditions (that is, an environment-independent DAE). To address the above problem, we propose two environment-dependent DAEs to reduce the influence of mismatches between training and test environments. In the first approach, we train various DAEs using speech from different acoustic environments, and the DAE for the condition that best matches the test condition is automatically selected (that is, a two-step environment-dependent DAE). To improve environment identification performance, we propose a DNN that uses both reverberant speech and estimated reverberation. In the second approach, we add estimated reverberation features to the input of the DAE (that is, a one-step environment-dependent DAE or a reverberation-aware DAE). The proposed method is evaluated using speech in simulated and real reverberant environments. Experimental results show that the environment-dependent DAE outperforms the environment-independent one in both simulated and real reverberant environments. For two-step environment-dependent DAE, the performance of environment identification based on the proposed DNN approach is also better than that of the conventional DNN approach, in which only reverberant speech is used and reverberation is not blindly estimated. And, the one-step environment-dependent DAE significantly outperforms the two-step environment-dependent DAE.

show abstract

Enhanced Speech Features by Single-Channel Joint Compensation of Noise and Reverberation

Cited by 44 publications

References 19 publications

Strategies for distant speech recognitionin reverberant environments

Strategies for distant speech recognitionin reverberant environments

Bayesian Feature Enhancement for Reverberation and Noise Robust Speech Recognition

Environment-dependent denoising autoencoder for distant-talking speech recognition

Contact Info

Product

Resources

About