Standard Feature Extraction Techniques Research Articles

COVID-19 classification using hybrid deep learning and standard feature extraction techniques

There is no doubt that COVID-19 disease rapidly spread all over the world, and effected the daily lives of all of the people. Nowadays, the reverse transcription polymerase chain reaction is the most way used to detect COVID-19 infection. Due to time consumed in this method and material limitation in the hospitals, there is a need for developing a robust decision support system depending on artificial intelligence (AI) techniques to recognize the infection at an early stage from a medical images. The main contribution in this research is to develop a robust hybrid feature extraction method for recognizing the COVID-19 infection. Firstly, we train the Alexnet on the images database and extract the first feature matrix. Then we used discrete wavelet transform (DWT) and principal component analysis (PCA) to extract the second feature matrix from the same images. After that, the desired feature matrices were merged. Finally, support vector machine (SVM) was used to classify the images. Training, validating, and testing of the proposed method were performed. Experimental results gave (97.6%, 98.5%) average accuracy rate on both chest X-ray and computed tomography (CT) images databases. The proposed hybrid method outperform a lot of standard methods and deep learning neural networks like Alexnet, Googlenet and other related methods.

Towards Robust Waveform-Based Acoustic Models

We study the problem of learning robust acoustic models in adverse environments, characterized by a significant mismatch between training and test conditions. This problem is of paramount importance for the deployment of speech recognition systems that need to perform well in unseen environments. First, we characterize data augmentation theoretically as an instance of vicinal risk minimization, which aims at improving risk estimates during training by replacing the delta functions that define the empirical density over the input space with an approximation of the marginal population density in the vicinity of the training samples. More specifically, we assume that local neighborhoods centered at training samples can be approximated using a mixture of Gaussians, and demonstrate theoretically that this can incorporate robust inductive bias into the learning process. We then specify the individual mixture components implicitly via data augmentation schemes, designed to address common sources of spurious correlations in acoustic models. To avoid potential confounding effects on robustness due to information loss, which has been associated with standard feature extraction techniques (e.g., FBANK and MFCC features), we focus on the waveform-based setting. Our empirical results show that the approach can generalize to unseen noise conditions, with 150% relative improvement in out-of-distribution generalization compared to training using the standard risk minimization principle. Moreover, the results demonstrate competitive performance relative to models learned using a training sample designed to match the acoustic conditions characteristic of test utterances.

Convolutional neural network-based automatic classification of midsagittal tongue gestural targets using B-mode ultrasound images.

Tongue gestural target classification is of great interest to researchers in the speech production field. Recently, deep convolutional neural networks (CNN) have shown superiority to standard feature extraction techniques in a variety of domains. In this letter, both CNN-based speaker-dependent and speaker-independent tongue gestural target classification experiments are conducted to classify tongue gestures during natural speech production. The CNN-based method achieves state-of-the-art performance, even though no pre-training of the CNN (with the exception of a data augmentation preprocessing) was carried out.

Auditory processing-based features for improving speech recognition in adverse acoustic conditions

The paper describes an auditory processing-based feature extraction strategy for robust speech recognition in environments, where conventional automatic speech recognition (ASR) approaches are not successful. It incorporates a combination of gammatone filtering, modulation spectrum and non-linearity for feature extraction in the recognition chain to improve robustness, more specifically the ASR in adverse acoustic conditions. The experimental results with standard Aurora-4 large vocabulary evaluation task revealed that the proposed features provide reliable and considerable improvement in terms of robustness in different noise conditions and are comparable to those of standard feature extraction techniques.

A dynamic automatic noisy speech recognition system for a single-channel hybrid noisy industrial environment

A dynamic noisy speech recognition system is developed to recognize single-channel speaker independent small spoken commands in a hybrid noisy industrial environment. The hybrid noise is environmental mixed noise distinguished as: (i) Strong, (ii) Time varying steady-unsteady, (iii) Mild. In [1] we presented a principal architecture for solving this problem. There were, however, several parts in the solution system missing. They are mainly on the technical level of speech recognition. These will be presented in this paper. A new adaptive feature extraction technique based on local trigonometric transformation (LTT) is introduced and examined. This is adapted with psychoacoustic quantities such as : a) Bark scaled critical band spectrum, b) loudness scale, and c) perceptual entropy. Here the spectral analysis is done by rising cut-off function, folding operation and discrete cosine transformation (DCT-IV) instead of Fourier transform. Then inverse DCT-IV and unfolding operation result in perceptual LTT (PLTT) features. These are classified by Gaussian mixture model (GMM) and recognized by hidden Markov model (HMM). The new PLTT features are more efficient and perceptually meaningful than the standard feature extraction techniques. The DANSR system is a novel solution for small commands to a long existing hybrid noise problem.

Discriminative Wavelet Packet Filter Bank Selection for Pattern Recognition

This paper addresses the problem of discriminative wavelet packet (WP) filter bank selection for pattern recognition. The problem is formulated as a complexity regularized optimization criterion, where the tree-indexed structure of the WP bases is explored to find conditions for reducing this criterion to a type of minimum cost tree pruning, a method well understood in regression and classification trees (CART). For estimating the conditional mutual information, adopted to compute the fidelity criterion of the minimum cost tree-pruning problem, a nonparametric approach based on product adaptive partitions is proposed, extending the Darbellay-Vajda data-dependent partition algorithm. Finally, experimental evaluation within an automatic speech recognition (ASR) task shows that proposed solutions for the WP decomposition problem are consistent with well understood empirically determined acoustic features, and the derived feature representations yield competitive performances with respect to standard feature extraction techniques.

Applying Sparse KPCA for Feature Extraction in Speech Recognition

This paper presents an analysis of the applicability of Sparse Kernel Principal Component Analysis (SKPCA) for feature extraction in speech recognition, as well as, a proposed approach to make the SKPCA technique realizable for a large amount of training data, which is an usual context in speech recognition systems. Although the KPCA (Kernel Principal Component Analysis) has proved to be an efficient technique for being applied to speech recognition, it has the disadvantage of requiring training data reduction, when its amount is excessively large. This data reduction is important to avoid computational unfeasibility and/or an extremely high computational burden related to the feature representation step of the training and the test data evaluations. The standard approach to perform this data reduction is to randomly choose frames from the original data set, which does not necessarily provide a good statistical representation of the original data set. In order to solve this problem a likelihood related re-estimation procedure was applied to the KPCA framework, thus creating the SKPCA, which nevertheless is not realizable for large training databases. The proposed approach consists in clustering the training data and applying to these clusters a SKPCA like data reduction technique generating the reduced data clusters. These reduced data clusters are merged and reduced in a recursive procedure until just one cluster is obtained, making the SKPCA approach realizable for a large amount of training data. The experimental results show the efficiency of SKPCA technique with the proposed approach over the KPCA with the standard sparse solution using randomly chosen frames and the standard feature extraction techniques.