Kalman filter for speech enhancement in cocktail party scenarios using a codebook-based approach

Kavalekalam, Mathew Shaji; Christensen, Mads Græsbøll; Gran, Fredrik; Boldt, Jesper B.

doi:10.1109/icassp.2016.7471663

Cited by 18 publications

(25 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The proposed method is based on an additive noise model assuming the speech and noise are statistically uncorrelated from [16], [17], i.e.,…”

Section: A Signal Modelmentioning

confidence: 99%

“…These parameters are estimated using a priori information from a trained codebook about the speech and noise spectral shapes in the form of LPC based on the approach in [16], [18], [17], where more details on the derivation of this method can be found. Given the observed vector of noisy samples y = [ y(0) y(1) .…”

Section: B Step 1: Estimate Parametersmentioning

confidence: 99%

“…The relevant features of the clean speech are based on the spectral envelope of the speech, which has been shown to be an important cue for speech intelligibility [15]. The spectral envelopes of the clean speech and the noise signal are estimated as the most suitable combination from a predefined speech and noise spectra dictionary, a codebook, which best fits the noisy speech signal using a codebookbased approach [16], [17]. These codebooks consist of filter coefficients that capture the overall structure of the spectral envelope.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Non-intrusive intelligibility prediction using a codebook-based approach

Sørensen

Kavalekalam

Xenaki³

et al. 2017

2017 25th European Signal Processing Conference (EUSIPCO)

Self Cite

View full text Add to dashboard Cite

Abstract-It could be beneficial for users of hearing aids if these were able to automatically adjust the processing according to the speech intelligibility in the specific acoustic environment. Most speech intelligibility metrics are intrusive, i.e., they require a clean reference signal, which is rarely available in real-life applications. This paper proposes a method, which allows using an intrusive short-time objective intelligibility (STOI) metric without requiring access to a clean signal. The clean speech reference signal is replaced by the clean speech envelope spectrum estimated from the noisy signal. The spectral envelope has been shown to be an important cue for speech intelligibility and is used as the reference signal inside STOI. The spectral envelopes are estimated as a combination of predefined dictionaries, i.e., codebooks, that best fits the noisy speech signal. The simulations show a high correlation between the proposed non-intrusive codebookbased STOI (NIC-STOI) and the intrusive STOI indicating that NIC-STOI is a suitable metric for automatic classification of speech signals.

show abstract

“…The proposed method is based on an additive noise model assuming the speech and noise are statistically uncorrelated from [16], [17], i.e.,…”

Section: A Signal Modelmentioning

confidence: 99%

Section: B Step 1: Estimate Parametersmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Non-intrusive intelligibility prediction using a codebook-based approach

Sørensen

Kavalekalam

Xenaki³

et al. 2017

2017 25th European Signal Processing Conference (EUSIPCO)

Self Cite

View full text Add to dashboard Cite

show abstract

“…In Kalman filter, the dynamics of the signal generation process is being modeled by state equation. The noisy and distorted components will be modeled by observation equations [6][7][8], so it acts as recursive data processing algorithm. Hence it can be used for cancelling stationary and nonstatioinary noise.…”

Section: Kalman Filtermentioning

confidence: 99%

“…Over many decades researchers are focused in this area and developed the different algorithms to remove the noise which is present along with the speech signal. [4][5][6][7][8] To suppress the backhgrount noise still adaptive filter is better tool. In the basic adaptive filter like least mean square (LMS) algorithm the step size remains constant in updating filter coefficient equation for all the input samples [1,2].…”

Section: Introductionmentioning

confidence: 99%

System Identification in Speech Signal Using Modified Kalman Based Normalized Least Mean Square (NLMS) Algorithm

Rathnakara¹

2017

IJARCSSE

View full text Add to dashboard Cite

Abstract-In this paper a modified Kalman based normalized least mean square algorithm is proposed for system identification. The performance of the proposed algorithm is compared with normalized least mean square (NLMS) and original Kalman based normalized least mean square (KLMS) algorithm using standard IEEE sentence (SP23) of NOIZEUS database with different types of real world noises at different SNR. The proposed algorithm shows better output SNR and speed of convergence compared with NLMS and KLMS algorithm

show abstract

A hybrid speech enhancement system with DNN based speech reconstruction and Kalman filtering

Zhu

Ouyang

et al. 2020

Multimed Tools Appl

View full text Add to dashboard Cite

Kalman filter for speech enhancement in cocktail party scenarios using a codebook-based approach

Cited by 18 publications

References 19 publications

Non-intrusive intelligibility prediction using a codebook-based approach

Non-intrusive intelligibility prediction using a codebook-based approach

System Identification in Speech Signal Using Modified Kalman Based Normalized Least Mean Square (NLMS) Algorithm

A hybrid speech enhancement system with DNN based speech reconstruction and Kalman filtering

Contact Info

Product

Resources

About