scholarly journals Assessment of Speech Quality during Speech Rehabilitation Based on the Solution of the Classification Problem

2021 ◽  
Author(s):  
Evgeny Kostyuchenko ◽  
Ivan Rakhmanenko ◽  
Alexander Shelupanov ◽  
Lidiya Balatskaya ◽  
Ivan Sidorov
Keyword(s):  
Neural Network ◽  
Classification Problem ◽  
Speech Quality ◽  
Speech Signals ◽  
Reference Class ◽  
Experimental Assessment ◽  
Speech Rehabilitation ◽  
Expert Assessments

The article considers an approach to the problem of assessing the quality of speech during speech rehabilitation as a classification problem. For this, a classifier is built on the basis of an LSTM neural network for dividing speech signals into two classes: before the operation and immediately after. At the same time, speech before the operation is the standard to which it is necessary to approach in the process of rehabilitation. The metric of belonging of the evaluated signal to the reference class acts as an assessment of speech. An experimental assessment of rehabilitation sessions and a comparison of the resulting assessments with expert assessments of phrasal intelligibility were carried out.

scholarly journals Speech Enhancement for Secure Communication Using Coupled Spectral Subtraction and Wiener Filter

Electronics ◽  
2019 ◽  
Vol 8 (8) ◽  
pp. 897 ◽  
Author(s):  
Hilman Pardede ◽  
Kalamullah Ramli ◽  
Yohan Suryanto ◽  
Nur Hayati ◽  
Alfan Presekal
Keyword(s):  
Speech Enhancement ◽  
Communication System ◽  
Secure Communication ◽  
Wiener Filter ◽  
Speech Quality ◽  
Spectral Subtraction ◽  
Speech Signals ◽  
Voice Activity Detector ◽  
Noise Estimate

The encryption process for secure voice communication may degrade the speech quality when it is applied to the speech signals before encoding them through a conventional communication system such as GSM or radio trunking. This is because the encryption process usually includes a randomization of the speech signals, and hence, when the speech is decrypted, it may perceptibly be distorted, so satisfactory speech quality for communication is not achieved. To deal with this, we could apply a speech enhancement method to improve the quality of decrypted speech. However, many speech enhancement methods work by assuming noise is present all the time, so the voice activity detector (VAD) is applied to detect the non-speech period to update the noise estimate. Unfortunately, this assumption is not valid for the decrypted speech. Since the encryption process is applied only when speech is detected, distortions from the secure communication system are characteristically different. They exist when speech is present. Therefore, a noise estimator that is able to update noise even when speech is present is needed. However, most noise estimator techniques only adapt to slow changes of noise to avoid over-estimation of noise, making them unsuitable for this task. In this paper, we propose a speech enhancement technique to improve the quality of speech from secure communication. We use a combination of the Wiener filter and spectral subtraction for the noise estimator, so our method is better at tracking fast changes of noise without over-estimating them. Our experimental results on various communication channels indicate that our method is better than other popular noise estimators and speech enhancement methods.

scholarly journals Early Stopping Criteria for Levenberg-Marquardt Based Neural Network Training Optimization

2018 ◽  
Vol 7 (4.36) ◽  
pp. 1194
Author(s):  
Azizah Suliman ◽  
Batyrkhan Omarov
Keyword(s):  
Neural Network ◽  
Classification Problem ◽  
Correct Choice ◽  
Network Training ◽  
The Neural Network ◽  
Marquardt Algorithm ◽  
Levenberg Marquardt ◽  
Stop Condition ◽  
Hidden Neurons

In this research we train a direct distributed neural network using Levenberg-Marquardt algorithm. In order to prevent overtraining, we proposed correctly recognized image percentage based on early stop condition and conduct the experiments with different stop thresholds for image classification problem. Experiment results show that the best early stop condition is 93% and other increase in stop threshold can lead to decrease in the quality of the neural network. The correct choice of early stop condition can prevent overtraining which led to the training of a neural network with considerable number of hidden neurons.  

scholarly journals Evaluation of Speech Quality Through Recognition and Classification of Phonemes

Symmetry ◽  
2019 ◽  
Vol 11 (12) ◽  
pp. 1447
Author(s):  
Svetlana Pekarskikh ◽  
Evgeny Kostyuchenko ◽  
Lidiya Balatskaya
Keyword(s):  
Neural Network ◽  
Surgical Treatment ◽  
Speech Intelligibility ◽  
Vocal Tract ◽  
Main Idea ◽  
Speech Quality ◽  
Training Dataset ◽  
Speech Rehabilitation ◽  
The Neural Network ◽  
Before And After

This paper discusses an approach for assessing the quality of speech while undergoing speech rehabilitation. One of the main reasons for speech quality decrease during the surgical treatment of vocal tract diseases is the loss of the vocal tractˈs parts and the disruption of its symmetry. In particular, one of the most common oncological diseases of the oral cavity is cancer of the tongue. During surgical treatment, a glossectomy is performed, which leads to the need for speech rehabilitation to eliminate the occurring speech defects, leading to a decrease in speech intelligibility. In this paper, we present an automated approach for conducting the speech quality evaluation. The approach relies on a convolutional neural network (CNN). The main idea of the approach is to train an individual neural network for a patient before having an operation to recognize typical sounding of phonemes for their speech. The neural network will thereby be able to evaluate the similarity between the patientˈs speech before and after the surgery. The recognition based on the full phoneme set and the recognition by groups of phonemes were considered. The correspondence of assessments obtained through the autorecognition approach with those from the human-based approach is shown. The automated approach is principally applicable to defining boundaries between phonemes. The paper shows that iterative training of the neural network and continuous updating of the training dataset gradually improve the ability of the CNN to define boundaries between different phonemes.

Convolutional Neural Network for Customer’s Opinion on Amazon Products

2019 ◽  
Vol 8 (3) ◽  
pp. 6634-6643 ◽  
Keyword(s):  
Neural Network ◽  
Convolutional Neural Network ◽  
Sentiment Analysis ◽  
Latent Dirichlet Allocation ◽  
Opinion Mining ◽  
Text Documents ◽  
Customer Churn ◽  
Learning Classifier ◽  
Review Spam

Opinion mining and sentiment analysis are valuable to extract the useful subjective information out of text documents. Predicting the customer’s opinion on amazon products has several benefits like reducing customer churn, agent monitoring, handling multiple customers, tracking overall customer satisfaction, quick escalations, and upselling opportunities. However, performing sentiment analysis is a challenging task for the researchers in order to find the users sentiments from the large datasets, because of its unstructured nature, slangs, misspells and abbreviations. To address this problem, a new proposed system is developed in this research study. Here, the proposed system comprises of four major phases; data collection, pre-processing, key word extraction, and classification. Initially, the input data were collected from the dataset: amazon customer review. After collecting the data, preprocessing was carried-out for enhancing the quality of collected data. The pre-processing phase comprises of three systems; lemmatization, review spam detection, and removal of stop-words and URLs. Then, an effective topic modelling approach Latent Dirichlet Allocation (LDA) along with modified Possibilistic Fuzzy C-Means (PFCM) was applied to extract the keywords and also helps in identifying the concerned topics. The extracted keywords were classified into three forms (positive, negative and neutral) by applying an effective machine learning classifier: Convolutional Neural Network (CNN). The experimental outcome showed that the proposed system enhanced the accuracy in sentiment analysis up to 6-20% related to the existing systems.

scholarly journals Nodule Detection with Convolutional Neural Network Using Apache Spark and GPU Frameworks

2021 ◽  
Vol 11 (6) ◽  
pp. 2838
Author(s):  
Nikitha Johnsirani Venkatesan ◽  
Dong Ryeol Shin ◽  
Choon Sung Nam
Keyword(s):  
Neural Network ◽  
Radiation Dose ◽  
Convolutional Neural Network ◽  
Model Performance ◽  
Performance Comparison ◽  
Apache Spark ◽  
Training Time ◽  
Learning Framework ◽  
Proposed Model

In the pharmaceutical field, early detection of lung nodules is indispensable for increasing patient survival. We can enhance the quality of the medical images by intensifying the radiation dose. High radiation dose provokes cancer, which forces experts to use limited radiation. Using abrupt radiation generates noise in CT scans. We propose an optimal Convolutional Neural Network model in which Gaussian noise is removed for better classification and increased training accuracy. Experimental demonstration on the LUNA16 dataset of size 160 GB shows that our proposed method exhibit superior results. Classification accuracy, specificity, sensitivity, Precision, Recall, F1 measurement, and area under the ROC curve (AUC) of the model performance are taken as evaluation metrics. We conducted a performance comparison of our proposed model on numerous platforms, like Apache Spark, GPU, and CPU, to depreciate the training time without compromising the accuracy percentage. Our results show that Apache Spark, integrated with a deep learning framework, is suitable for parallel training computation with high accuracy.

scholarly journals Hybrid Reference Current Generation Theory for Solar Fed UPFC System

Energies ◽  
2021 ◽  
Vol 14 (6) ◽  
pp. 1527
Author(s):  
R. Senthil Kumar ◽  
K. Mohana Sundaram ◽  
K. S. Tamilselvan
Keyword(s):  
Neural Network ◽  
Power Factor ◽  
Reactive Power ◽  
Voltage Source ◽  
Unity Power Factor ◽  
Current Generation ◽  
Load Voltage ◽  
Reference Current Generation ◽  
Source Current

The extensive usage of power electronic components creates harmonics in the voltage and current, because of which, the quality of delivered power gets affected. Therefore, it is essential to improve the quality of power, as we reveal in this paper. The problems of load voltage, source current, and power factors are mitigated by utilizing the unified power flow controller (UPFC), in which a combination of series and shunt converters are combined through a DC-link capacitor. To retain the link voltage and to maximize the delivered power, a PV module is introduced with a high gain converter, named the switched clamped diode boost (SCDB) converter, in which the grey wolf optimization (GWO) algorithm is instigated for tracking the maximum power. To retain the link-voltage of the capacitor, the artificial neural network (ANN) is implemented. A proper control of UPFC is highly essential, which is achieved by the reference current generation with the aid of a hybrid algorithm. A genetic algorithm, hybridized with the radial basis function neural network (RBFNN), is utilized for the generation of a switching sequence, and the generated pulse has been given to both the series and shunt converters through the PWM generator. Thus, the source current and load voltage harmonics are mitigated with reactive power compensation, which results in attaining a unity power factor. The projected methodology is simulated by MATLAB and it is perceived that the total harmonic distortion (THD) of 0.84% is attained, with almost a unity power factor, and this is validated with FPGA Spartan 6E hardware.

scholarly journals AI-based localization and classification of skin disease with erythema

2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Ha Min Son ◽  
Wooho Jeon ◽  
Jinhyun Kim ◽  
Chan Yeong Heo ◽  
Hye Jin Yoon ◽  
...  
Keyword(s):  
Neural Network ◽  
Skin Diseases ◽  
Classification Model ◽  
Screening Tests ◽  
Sensitivity Score ◽  
Common Skin ◽  
Novel Method ◽  
Improved Performance ◽  
High Level

AbstractAlthough computer-aided diagnosis (CAD) is used to improve the quality of diagnosis in various medical fields such as mammography and colonography, it is not used in dermatology, where noninvasive screening tests are performed only with the naked eye, and avoidable inaccuracies may exist. This study shows that CAD may also be a viable option in dermatology by presenting a novel method to sequentially combine accurate segmentation and classification models. Given an image of the skin, we decompose the image to normalize and extract high-level features. Using a neural network-based segmentation model to create a segmented map of the image, we then cluster sections of abnormal skin and pass this information to a classification model. We classify each cluster into different common skin diseases using another neural network model. Our segmentation model achieves better performance compared to previous studies, and also achieves a near-perfect sensitivity score in unfavorable conditions. Our classification model is more accurate than a baseline model trained without segmentation, while also being able to classify multiple diseases within a single image. This improved performance may be sufficient to use CAD in the field of dermatology.

Sign in / Sign up

Export Citation Format

Share Document