Wissam A. Jassim

New orthogonal polynomials for speech signal and image processing

This study introduces a new set of orthogonal polynomials and moments and the sets application in signal and image processing. This polynomial is derived from two well-known orthogonal polynomials: the Tchebichef and Krawtchouk polynomials. This study attempts to present the following: (i) the mathematical and theoretical frameworks for the definition of this polynomial including the modelling of signals with the various analytical properties it contains, as well as, recurrence relations and transform equations that need to be addressed; and (ii) the results of empirical tests that compare the representational capabilities of this polynomial with those of the more traditional Tchebichef and Krawtchouk polynomials using speech and image signals from different databases. This study attempts to demonstrate that the proposed polynomials can be applied in the field of signal and image processing because of the promising properties of this polynomial especially in its localisation and energy compaction capabilities.

Prediction of speech intelligibility using a neurogram orthogonal polynomial measure (NOPM)

Sensorineural hearing loss (SNHL) is an increasingly prevalent condition, resulting from damage to the inner ear and causing a reduction in speech intelligibility. This paper proposes a new speech intelligibility prediction metric, the neurogram orthogonal polynomial measure (NOPM). This metric applies orthogonal moments to the auditory neurogram to predict speech intelligibility for listeners with and without hearing loss. The model simulates the responses of auditory-nerve fibers to speech signals under quiet and noisy conditions. Neurograms were created using a physiologically based computational model of the auditory periphery. A well-known orthogonal polynomial measure, Krawtchouk moments, was applied to extract features from the auditory neurogram. The predicted intelligibility scores were compared to subjective results, and NOPM showed a good fit with the subjective scores for normal listeners and also for listeners with hearing loss. The proposed metric has a realistic and wider dynamic range than corresponding existing metrics, such as mean structural similarity index measure and neurogram similarity index measure, and the predicted scores are also well-separated as a function of hearing loss. The application of this metric could be extended for assessing hearing-aid and speech-enhancement algorithms.

Enhancing noisy speech signals using orthogonal moments

This study describes a new approach to enhance noisy speech signals using the discrete Tchebichef transform (DTT) and the discrete Krawtchouk transform (DKT). The DTT and DKT are based on well-known orthogonal moments: the Tchebichef and Krawtchouk moments, respectively. The representations of speech signals using a limited number of moment coefficients and their behaviour in the domain of orthogonal moments are shown. The method involves removing noise from the signal using a minimum-mean-square error in the domain of the DTT or DKT. According to comparisons with traditional methods, the initial experiments yield promising results and show that orthogonal moments are applicable in the field of speech signal enhancement. The application of orthogonal moments could be extended to speech analysis, compression and recognition.

Face Recognition Using Discrete Tchebichef-Krawtchouk Transform

In this paper, a face recognition system based on Discrete Tchebichef-Krawtchouk Transform DTKT and Support Vector Machines SVMs is proposed. The objective of this paper is to present the following: (1) the mathematical and theoretical frameworks for the definition of the DTKT including transform equations that need to be addressed. (2) the DTKT features used in the classification of faces. (3) results of empirical tests that compare the representational capabilities of this transform with other types of discrete transforms such as Discrete Tchebichef transform DTT, discrete Krawtchouk Transform DKT, and Discrete Cosine transform DCT. The system is tested on a large number of faces collected from ORL and Yale face databases. Empirical results show that the proposed transform gives very good overall accuracy under clean and noisy conditions.

A Robust Speaker Identification System Using the Responses from a Model of the Auditory Periphery

Speaker identification under noisy conditions is one of the challenging topics in the field of speech processing applications. Motivated by the fact that the neural responses are robust against noise, this paper proposes a new speaker identification system using 2-D neurograms constructed from the responses of a physiologically-based computational model of the auditory periphery. The responses of auditory-nerve fibers for a wide range of characteristic frequency were simulated to speech signals to construct neurograms. The neurogram coefficients were trained using the well-known Gaussian mixture model-universal background model classification technique to generate an identity model for each speaker. In this study, three text-independent and one text-dependent speaker databases were employed to test the identification performance of the proposed method. Also, the robustness of the proposed method was investigated using speech signals distorted by three types of noise such as the white Gaussian, pink, and street noises with different signal-to-noise ratios. The identification results of the proposed neural-response-based method were compared to the performances of the traditional speaker identification methods using features such as the Mel-frequency cepstral coefficients, Gamma-tone frequency cepstral coefficients and frequency domain linear prediction. Although the classification accuracy achieved by the proposed method was comparable to the performance of those traditional techniques in quiet, the new feature was found to provide lower error rates of classification under noisy environments.

Reference-Free Assessment of Speech Intelligibility Using Bispectrum of an Auditory Neurogram

Sensorineural hearing loss occurs due to damage to the inner and outer hair cells of the peripheral auditory system. Hearing loss can cause decreases in audibility, dynamic range, frequency and temporal resolution of the auditory system, and all of these effects are known to affect speech intelligibility. In this study, a new reference-free speech intelligibility metric is proposed using 2-D neurograms constructed from the output of a computational model of the auditory periphery. The responses of the auditory-nerve fibers with a wide range of characteristic frequencies were simulated to construct neurograms. The features of the neurograms were extracted using third-order statistics referred to as bispectrum. The phase coupling of neurogram bispectrum provides a unique insight for the presence (or deficit) of supra-threshold nonlinearities beyond audibility for listeners with normal hearing (or hearing loss). The speech intelligibility scores predicted by the proposed method were compared to the behavioral scores for listeners with normal hearing and hearing loss both in quiet and under noisy background conditions. The results were also compared to the performance of some existing methods. The predicted results showed a good fit with a small error suggesting that the subjective scores can be estimated reliably using the proposed neural-response-based metric. The proposed metric also had a wide dynamic range, and the predicted scores were well-separated as a function of hearing loss. The proposed metric successfully captures the effects of hearing loss and supra-threshold nonlinearities on speech intelligibility. This metric could be applied to evaluate the performance of various speech-processing algorithms designed for hearing aids and cochlear implants.

Blind image quality assessment for Gaussian blur images using exact Zernike moments and gradient magnitude

Abstract Features that exhibit human perception on the effect of blurring on digital images are useful in constructing a blur image quality metric. In this paper, we show some of the exact Zernike moments (EZMs) that closely model the human quality scores for images of varying degrees of blurriness can be used to measure these distortions. A theoretical framework is developed to identify these EZMs. Together with the selected EZMs, the gradient magnitude (GM), which measures the contrast information, is used as a weight in the formulation of the proposed blur metric. The design of the proposed metric consists of two stages. In the first stage, the EZM differences and the GM dissimilarities between the edge points of the test image and the same re-blurred image are extracted. Next, the mean of the weighted EZM features are then pooled to produce a quality score using support vector machine regressor (SVR). We compare the performance of the proposed blur metric with other state-of-the-art full-reference (FR) and no-reference (NR) blur metrics on three benchmark databases. The results using Pearson׳s correlation coefficient (CC) and Spearman׳s ranked-order correlation coefficient (SROCC) for the LIVE image database are 0.9659 and 0.9625 respectively. Similarly, high correlations with the subjective scores are achieved for the other two databases as well.

Speech signals representation by Discrete Transforms

In this paper, an attempt was made to analyze the speech reconstruction accuracy when using different basis functions as the kernel for a reversible transform. Various transforms such as Discrete Cosine Transform DCT, Discrete Tchebichef Transform DTT, Ordered Hadamard Transform, and Discrete Haar Transform, are defined and examined. We have found that the DCT and DTT transforms have provided the greatest energy compactness properties for noise free speech sets. While, for noisy speech signals, DCT and Haar transform have the best signal representations in the transform domain as shown in the simulation results section.

Speech quality assessment using 2D neurogram orthogonal moments

This study proposes a new objective speech quality measure using the responses of a physiologically-based computational model of auditory nerve (AN). The population response of the model AN fibers to a speech signal is represented by a 2D neurogram, and features of the neurogram are extracted by orthogonal moments. A special type of orthogonal moment, the orthogonal Tchebichef-Krawtchouk moment, is used in this study. The proposed measure is compared to the subjective scores from two standard databases, the NOIZEUS and the supplement 23 to the P series (P.Sup23) of ITU-T Recommendations. The NOIZEUS database is used in the assessment of 11 speech enhancement algorithms whereas the P.Sup23 database is used in the ITU-T 8źkbit/s codec (Recommendation G.729) characterization test. The performance of the proposed speech quality measure is also compared to the results from some traditional objective quality measures. In general, the proposed neural-response-based metric yielded better results than most of the traditional acoustic-property-based quality measures. The proposed metric can be applied to evaluate the performance of various speech-enhancement algorithms and compression systems.

Neural response based phoneme classification under noisy condition

Human listeners are capable of recognizing speech in noisy environment, while most of the traditional speech recognition methods do not perform well in the presence of noise. Unlike traditional Mel-frequency cepstral coefficient (MFCC)-based method, this study proposes a phoneme classification technique using the neural responses of a physiologically-based computational model of the auditory periphery. Neurograms were constructed from the responses of the model auditory nerve to speech phonemes. The features of neurograms were used to train the recognition system using a Gaussian Mixture Model (GMM) classification technique. Performance was evaluated for different types of phonemes such as stops, fricatives and vowels from the TIMIT database for both under quiet and noisy conditions. Although performance of the proposed method is comparable with that of MFCC-based classifier in quiet condition, the neural response-based proposed method outperforms the traditional MFCC-based method under noisy conditions even with the use of less number of features in the proposed method. The proposed method could be used in the field of speech recognition such as speech to text application, especially under noisy conditions.