2D Psychoacoustic modeling of equivalent masking for automatic speech recognition

Dai, Peng; Rudzicz, Frank; Soon, Ing Yann; Mihailidis, Alex; Ding, Huijun

doi:10.1016/j.sigpro.2015.03.010

Cited by 6 publications

(3 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In order to evaluate the effectiveness of the proposed algorithm, several comparisons are made against the MFCC baseline [1], RASATA-PLP [4], MFCC+MVA [10] and minimum mean-square error (MMSE) spectral amplitude estimator [15]. We added to our comparison two newly effective techniques [27,28], the first technique denoted (TECH1) proposes the implementation of the 2D psychoacoustic models to MFCC features and the second technique denoted (TECH2) investigates the distribution of Mel-filtered log-spectrum of speech signals in noisy environments.…”

Section: Comparison With Other Methodsmentioning

confidence: 99%

Feature Extraction Algorithm Using New Cepstral Techniques for Robust Speech Recognition

Korba

Bourouba

Djemili

2020

MJCS

View full text Add to dashboard Cite

In this work, we propose a novel feature extraction algorithm that improves the robustness of automatic speech recognition (ASR) systems in the presence of various types of noise. The proposed algorithm uses a new cepstral technique based on the differential power spectrum (DPS) instead of the power spectrum (PS), the algorithm replaces the logarithmic non linearity by the power function. In order to reduce cepstral coefficients mismatches between training and testing conditions, we used the mean and variance normalization, then we apply auto-regression movingaverage filtering (MVA) in the cepstral domain. The ASR experiments were conducted using two databases, the first is LASA digit database designed for recognition the isolated Arabic digits in the presence of different types of noise. The second is Aurora 2 noisy speech database designed to recognize connected English digits in various operating environments. The experimental results show a substantial improvement from the proposed algorithm over the baseline Mel Frequency Cepstral Coefficients (MFCC), the relative improvement is the 28.92% for LASA database and is the 44.43% for aurora 2 database. The performance of our proposed algorithm was tested and verified by extensive comparisons with the state-of-the-art noise-robust features in aurora 2.

show abstract

Section: Comparison With Other Methodsmentioning

confidence: 99%

Feature Extraction Algorithm Using New Cepstral Techniques for Robust Speech Recognition

Korba

Bourouba

Djemili

2020

MJCS

View full text Add to dashboard Cite

show abstract

“…Speech recognition is becoming effective that yielding a result of over 90% and above word level accuracy for vocabulary speech recognition tasks. However, the accuracy of speech recognition may drop to less than 85% with the effect of noise [2]. Nowadays, this system is very popular in technology where it is applied in medicine, telephone network [3], security devices, ATM machines and computers [4].…”

Section: Introductionmentioning

confidence: 99%

Novel Statistical Clustering Method for Accurate Characterization of Word Pronunciation

Bahari

Musa

Nuawi

et al. 2017

International Journal on Advanced Science, Engineering and Information Technology

View full text Add to dashboard Cite

This paper discusses the development method to determine the accuracy of pronunciation of the word using global statistical signal analysis parameters. An engineering word that has been chosen is 'leaching'. The pronunciation of the word 'leaching' in the French language has been recorded from 1 native speaker and 4 students. The recording processes use a microphone-laptop system configuration and the signal analyzing processes use MATLAB software. Time and frequency domain plots show a variety of waveforms according to the recorded pronunciation. For data processing, statistical signal analysis parameters involved to extract the signal's features are kurtosis, root mean square and skewness. The mapping process has been performed to cluster each data. The position of the samples from the students is referred to the samples from the native speaker. The result of the accuracy of the pronunciation of words for each student can be evaluated through the comparison of the position of all the samples. In conclusion, the development of mapping and clustering methods are able to characterize the accuracy of the pronunciation of words.

show abstract

“…A signal is most likely to get masked by another signal with frequency components which are near to, or the same as, that of the signal. When masking event occurs among any two signals which appear at similar instant of time it is called simultaneous masking or spectral masking, on the other hand , signals which are comparatively postponed in time it is called temporal or non simultaneous masking [10].…”

Section: Introductionmentioning

confidence: 99%

Auditory Masking Removal for Perception Improvement of Hearing Impaired Persons

M¹,

A.²

2016

CAE

View full text Add to dashboard Cite

A novel method is projected in order to get a better speech perception for the hearing impaired people. The algorithm is developed to achieve a specific number of bands of the preferred speech material for dichotic presentation.These bands are obtained using a set of dyadic filters which gives sharp transitions between selected bands. For implementing this algorithm MATLAB software is used .At first by implementing the code eighteen bands of Constant Bandwidth, nineteen bands of 1/3 Octave bandwidths and eighteen bands of critical bandwidth using pair of complementary comb filters from the filter sets are obtained. Ten subjects of various age group participated for the listening test .The test material used for the testing purpose involves a set of 15 Vowel Consonant Vowel (VCV) context .The performed test outcome proves that the projected method is usefull for improving the recognition score and also reduces the response time.

show abstract

2D Psychoacoustic modeling of equivalent masking for automatic speech recognition

Cited by 6 publications

References 31 publications

Feature Extraction Algorithm Using New Cepstral Techniques for Robust Speech Recognition

Feature Extraction Algorithm Using New Cepstral Techniques for Robust Speech Recognition

Novel Statistical Clustering Method for Accurate Characterization of Word Pronunciation

Auditory Masking Removal for Perception Improvement of Hearing Impaired Persons

Contact Info

Product

Resources

About