VTS residual noise compensation

Segura, J.; Benítez, Domingo; Torre, De La; Dupont,; Rubio,

doi:10.1109/icassp.2002.1005763

Cited by 4 publications

(3 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, large amount of data is required to reliably estimate the joint histogram, which may not be available in practice. In most of the literature (Segura et al, 2002;Obuchi and Stern, 2003;de la Torre et al, 2005) equalization is done in the cepstral domain as the cepstral coefficients are approximately uncorrelated due to multiplication of discrete cosine transform of type 2 (DCT-2) (Victoria et al, 1995;Logan, 2000) to LMFB coefficients. Equalization in cepstral domain does not account for a band-specific equalization as the frequency band specific effects of noise get distributed among all the cepstral coefficients by the application of DCT transform to LMFB coefficients.…”

Section: Introductionmentioning

confidence: 99%

Sub-band based histogram equalization in cepstral domain for speech recognition

Joshi

Bilgi

Umesh

et al. 2015

Speech Communication

View full text Add to dashboard Cite

Section: Introductionmentioning

confidence: 99%

Sub-band based histogram equalization in cepstral domain for speech recognition

Joshi

Bilgi

Umesh

et al. 2015

Speech Communication

View full text Add to dashboard Cite

“…C. Residual noise removal After speech enhancement and feature enhancement by ALGONQUIN, there is also the global shift of the probability distributions of the features caused by the presence of noise. The expected value of the estimator is the true value of clean speech plus a varying bias which is a non-linear function of both the actual and estimated SNR [10].…”

Section: B Noise Compensationmentioning

confidence: 99%

“…E[x] = u, + log(i+- (10) where s -,U -/n is an estimate of the instantaneous SNR. From (10), we can find that there is a shift of x and the shift can be decreased by increasing SNR.…”

Section: B Noise Compensationmentioning

confidence: 99%

A Noise Robust Front-End Using Wiener Filter, Probability Model And CMS For ASR

Wang

et al.

2005 International Conference on Natural Language Processing and Knowledge Engineering

View full text Add to dashboard Cite

A novel and noise robust front-end based on the combination of spectral noise reduction and Probability Modelbased feature compensation and Cepstral Mean Subtraction (CMS) is proposed. Mel filter-bank outputs can be affected by additive noise primarily because of the vulnerable spectral valleys. An instantaneous Wiener filter is used to improve SNR of the spectral valley. Because the compensated MFCC is an approximation of the clean one and retains a residual mismatch, features are further processed by CMS in order to remove the global shift of the mean. In the presence of additive noise, ASR experiments reveal that a cascade fashion use of these techniques improves recognition performance greatly. For the 863 continuous Chinese speech databases, the average recognition rate across different noise types is improved from 34.63% (using unmodified MFCCs) to 84.63% (using the proposed techniques) at best.

show abstract

Robust Speech Activity Detection Using LDA Applied to FF Parameters

Padrell¹,

Macho²,

Nadeu³

Proceedings. (ICASSP '05). IEEE International Conference on Acoustics, Speech, and Signal Processing, 2005.

View full text Add to dashboard Cite

VTS residual noise compensation

Cited by 4 publications

References 0 publications

Sub-band based histogram equalization in cepstral domain for speech recognition

Sub-band based histogram equalization in cepstral domain for speech recognition

A Noise Robust Front-End Using Wiener Filter, Probability Model And CMS For ASR

Robust Speech Activity Detection Using LDA Applied to FF Parameters

Contact Info

Product

Resources

About