A data-driven approach to optimizing spectral speech enhancement methods for various error criteria

Erkelens, J.S.; Jensen, Jesper; Heusdens, Richard

doi:10.1016/j.specom.2006.06.012

Cited by 58 publications

(46 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We compared the speech quality obtained from the proposed residual gain estimation algorithm (denoted as RGE) with those of the conventional speech enhancement algorithm based on soft decision [3] (denoted as SESD) which was adopted in this work as the first stage module, and the data-driven approach proposed in [5] using the SNR grid look-up table (denoted as SGLT). For the test material, NOIZEUS corpus [9] was used which consisted of 30 IEEE sentences, and these speech data were corrupted by different types of additive noises at various SNR's.…”

Section: Resultsmentioning

confidence: 99%

“…The residual gain, which is estimated by a data-driven method, is applied to further adjust the gain obtained from the first stage. Experimental results show that the proposed algorithm performs better than both the conventional speech enhancement technique based on soft decision [3] and the data-driven approach using SNR grid look-up table [5].…”

Section: Discussionmentioning

confidence: 99%

“…Recently, several data-driven approaches to optimize speech enhancement methods have been proposed [5], [6]. In these approaches, the gain is obtained from an SNR lookup table trained in a variety of noisy conditions.…”

Section: Conventional Approaches To Speech Enhancementmentioning

confidence: 99%

“…A data-driven strategy is employed to estimate the residual gain, for which we use the SNR-related parameters provided by the first stage enhancement algorithm. From a number of experiments, we can see that the proposed residual gain estimation approach shows better performance compared with the soft decision based enhancement algorithm [3] which is used as the first stage module in this work and the data-driven approach proposed in [5].…”

Section: Introductionmentioning

confidence: 96%

See 3 more Smart Citations

Speech Enhancement Based on Data-Driven Residual Gain Estimation

Jin

Kim

Chang

2011

IEICE Trans. Inf. & Syst.

View full text Add to dashboard Cite

SUMMARYIn this letter, we propose a novel speech enhancement algorithm based on data-driven residual gain estimation. The entire system consists of two stages. At the first stage, a conventional speech enhancement algorithm enhances the input signal while estimating several signalto-noise ratio (SNR)-related parameters. The residual gain, which is estimated by a data-driven method, is applied to further enhance the signal at the second stage. A number of experimental results show that the proposed speech enhancement algorithm outperforms the conventional speech enhancement technique based on soft decision and the data-driven approach using SNR grid look-up table.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

Section: Conventional Approaches To Speech Enhancementmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 96%

See 2 more Smart Citations

Speech Enhancement Based on Data-Driven Residual Gain Estimation

Jin

Kim

Chang

2011

IEICE Trans. Inf. & Syst.

View full text Add to dashboard Cite

show abstract

“…The variance of YE is denoted as λE. The decision-directed approach [17] will be used for estimation of the prior SNRs, with a bias correction [18]. TheŶE(k, m) are transformed back into the time domain and the enhanced speech is formed by an overlap-add procedure.…”

Section: Speech Enhancementmentioning

confidence: 99%

Single-microphone late-reverberation suppression in noisy speech by exploiting long-term correlation in the DFT domain

Erkelens

Heusdens

2009

2009 IEEE International Conference on Acoustics, Speech and Signal Processing

Self Cite

View full text Add to dashboard Cite

We consider blind late-reverberation suppression in speech signals measured with a single microphone in noisy environments. We exploit that reverberant speech shows correlation over longer time spans than clean speech by predicting the contribution of reverberant energy to the current observed spectrum from the enhanced spectra of previous frames. The prediction parameters are recursively updated with estimates of the correlation coefficients between the current reverberant spectrum and enhanced previous spectra. The contributions of late reverberation and noise are suppressed by a standard noise reduction algorithm. The algorithm is shown to decrease the long-term correlation. It achieves significant improvements in segmental speech-to-interference ratio and Bark spectral distortion for typical reverberation times and noise levels, while almost no distortions are introduced in clean speech.

show abstract

Voice Activity Detection, Noise Estimation, and Adaptive Filters for Acoustic Signal Enhancement

Martin¹,

Kolossa²

2012

Techniques for Noise Robustness in Automatic Speech Recognition

View full text Add to dashboard Cite

A data-driven approach to optimizing spectral speech enhancement methods for various error criteria

Cited by 58 publications

References 14 publications

Speech Enhancement Based on Data-Driven Residual Gain Estimation

Speech Enhancement Based on Data-Driven Residual Gain Estimation

Single-microphone late-reverberation suppression in noisy speech by exploiting long-term correlation in the DFT domain

Voice Activity Detection, Noise Estimation, and Adaptive Filters for Acoustic Signal Enhancement

Contact Info

Product

Resources

About