Search results for: isolated word recognition

Authors: Heng Yang, Tao Luo, Yakun Zhang, Kai Wang, Wei Qin, Liang Xie, Ye Yan, Erwei Yin

Abstract:

Recent years have seen that audio-visual recognition has shown great potential in a strong noise environment. The existing method of audio-visual recognition has explored methods with ResNet and feature fusion. However, on the one hand, ResNet always occupies a large amount of memory resources, restricting the application in engineering. On the other hand, the feature merging also brings some interferences in a high noise environment. In order to solve the problems, we proposed an effective framework with bidirectional distillation. At first, in consideration of the good performance in extracting of features, we chose the light model, Efficientnet as our extractor of spatial features. Secondly, self-distillation was applied to learn more information from raw data. Finally, we proposed a bidirectional distillation in decision-level fusion. In more detail, our experimental results are based on a multi-model dataset from 24 volunteers. Eventually, the lipreading accuracy of our framework was increased by 2.3% compared with existing systems, and our framework made progress in audio-visual fusion in a high noise environment compared with the system of audio recognition without visual.

Keywords: lipreading, audio-visual, Efficientnet, distillation

Procedia PDF Downloads 110

Authors: Vadim E. Levit, Eugen Mandrescu

Abstract:

A graph is well-covered if all its maximal independent sets are of the same size. A well-covered graph is 1-well-covered if deletion of every vertex of the graph leaves it well-covered. It is known that a graph without isolated vertices is 1-well-covered if and only if every two disjoint independent sets are included in two disjoint maximum independent sets. Well-covered graphs are related to combinatorial commutative algebra (e.g., every Cohen-Macaulay graph is well-covered, while each Gorenstein graph without isolated vertices is 1-well-covered). Our intent is to construct several infinite families of 1-well-covered graphs using the following known graph operations: corona, join, and rooted product of graphs. Adopting some known techniques used to advantage for well-covered graphs, one can prove that: if the graph G has no isolated vertices, then the corona of G and H is 1-well-covered if and only if H is a complete graph of order two at least; the join of the graphs G and H is 1-well-covered if and only if G and H have the same independence number and both are 1-well-covered; if H satisfies the property that every three pairwise disjoint independent sets are included in three pairwise disjoint maximum independent sets, then the rooted product of G and H is 1-well-covered, for every graph G. These findings show not only how to generate some more families of 1-well-covered graphs, but also that, to this aim, sometimes, one may use graphs that are not necessarily 1-well-covered.

Keywords: maximum independent set, corona, concatenation, join, well-covered graph

Procedia PDF Downloads 179

Authors: Paulo Fernando Silva Filho, Elcio Hideiti Shiguemori

Abstract:

The selection of specific landmarks for an Unmanned Aerial Vehicles’ Visual Navigation systems based on Automatic Landmark Recognition has significant influence on the precision of the system’s estimated position. At the same time, manual selection of the landmarks does not guarantee a high recognition rate, which would also result on a poor precision. This work aims to develop an automatic landmark selection that will take the image of the flight area and identify the best landmarks to be recognized by the Visual Navigation Landmark Recognition System. The criterion to select a landmark is based on features detected by ORB or AKAZE and edges information on each possible landmark. Results have shown that disposition of possible landmarks is quite different from the human perception.

Keywords: clustering, edges, feature points, landmark selection, X-means

Procedia PDF Downloads 258

Authors: Marija Vitanova, Aleksandra Bogdanovic, Kemal Edip, Viktor Hristovski, Vlado Micov

Abstract:

Bridges represent critical structures in lifeline systems. They provide reliable modes of transportation, so their failure can seriously obstruct relief and rehabilitation work. Earthquake ground motions can cause significant damages in bridges, so during the strong earthquakes, they can easily collapse. The base isolation technique has been quite effective in seismic response mitigation of the bridges in reducing the piers base shear. The effect of soil structure interaction on the dynamic responses of seismically isolated three span girder bridge with viscous dampers is investigated. Viscous dampers are installed in the mid span of the bridge to control bearing displacement. The soil surrounding the foundation of piers has been analyzed by applying different soil densities in order to consider the soil stiffness. The soil medium has been assumed as a four layered infill as dense and loose medium. The boundaries in the soil medium are considered as infinite elements in order to absorb the radiating waves. The formulation of infinite elements is the same as for the finite elements in addition to the mapping of the domain. Based on the iso-parametric concept, the infinite element in global coordinate is mapped onto an element in local coordinate system. In the formulation of the infinite element, only the positive direction extends to infinity thus allowing the waves to propagate outside of the soil medium. Dynamic analyses for two levels of earthquake intensity are performed in time domain using direct integration method. In order to specify the effects of the SSI, the responses of the isolated and controlled isolated bridges are compared. It is observed that the soil surrounding the piers has significant effects on the bearing displacement of the isolated RC bridges. In addition, it is observed that the seismic responses of isolated RC bridge reduced significantly with the installation of the viscous dampers.

Keywords: viscous dampers, reinforced concrete girder bridges, seismic response, SSI

Procedia PDF Downloads 101

Authors: Alaaeldin Hamdy Ahmed Mohammed

Abstract:

Introduction: This study examines sports’ fans and non-interested public perception and recognition of the problems that have negative impacts upon the Egyptian sports, particularly football, through caricatures. Eight caricature paintings were designed to express eight problems affecting the Egyptian sports and its development. These paintings were distributed on two groups of the fans and the non-interested public. Methods: The study was limited to eight caricatures representing the eight issues which are: the impact of stopping the sports activity on athletes, the effect of clubs’ disagreement, fanaticism between the members of the ultras of different clubs, the negative impact of the mingling of politics into sports, the negative role of the clubs affects the professionalism of the promising players, the conflict between the national organization responsible for sports, the breaking in of the fans to the playgrounds, the impact of the lack of planning on the national team. The Results: The results showed that both sports fans and those who are not interested in sports recognized the problems that the caricatures refer to and criticizes exaggeration although the rate was higher for the fans. These caricatures contributed also in their recognition of the danger of the negative impact of these problems on the Egyptian sports, particularly football which is the most common at the Egyptian sports fans. Discussion: This finding echoes the conclusion that caricatures are distinctive in the adults’ facial stimuli that are either systematically exaggerated recognition of them.

Keywords: caricature, fans, football, sports

Procedia PDF Downloads 296

Authors: Bowen Ding, Yihao Kuang

Abstract:

In recent years, large-scale pre-trained language models have achieved state-of-the-art performance on a variety of natural language processing tasks. The word vectors produced by these language models can be viewed as dense encoded presentations of natural language that in text form. However, it is unknown how much linguistic information is encoded and how. In this paper, we construct several corresponding probing tasks for multiple linguistic information to clarify the encoding capabilities of different language models and performed a visual display. We firstly obtain word presentations in vector form from different language models, including BERT, ELMo, RoBERTa and GPT. Classifiers with a small scale of parameters and unsupervised tasks are then applied on these word vectors to discriminate their capability to encode corresponding linguistic information. The constructed probe tasks contain both semantic and syntactic aspects. The semantic aspect includes the ability of the model to understand semantic entities such as numbers, time, and characters, and the grammatical aspect includes the ability of the language model to understand grammatical structures such as dependency relationships and reference relationships. We also compare encoding capabilities of different layers in the same language model to infer how linguistic information is encoded in the model.

Keywords: language models, probing task, text presentation, linguistic information

Procedia PDF Downloads 83

Authors: Madhavi

Abstract:

All linguistic units are context-dependent. They occur in particular settings, from which they derive much of their import, and are recognized by speakers as distinct entities only through a process of abstraction. Most of the words have several concepts associated with them and convey a number of meanings in different contexts in any language. For instance, there are different uses of the word good as an adjective from English. The adjective good expresses many senses like (1) ‘high quality of someone or something’ (2) ‘efficient’ (3) ‘virtuous’ (4) ‘reliable’ etc. These senses will be analyzed by using cognitive semantics framework. The context has the power to insulate one meaning from all the other meanings in communication. This paper will provide a cognitive semantic analysis. The basic tenet of cognitive semantics is the sense of a word is the way we conceptualize it. Our conceptualization is based on the physical experience we go through. Cognitive semantics tries to capture this conceptualization in terms of some categories like schema, frame, and domain. Cognitive semantics is a subfield of cognitive linguistics. Cognitive linguistics studies the language creation, learning, and usage by the reference to human cognition. The semantic structure is conceptual structure which is related to the concepts which are the elements of reason and constitute the meanings of words and linguistic expressions. Cognitive semantics studies how our mind works for the meaning of any word and how it perceives meaning from the environment through senses and works to map with the knowledge which already exists in our mind through experience. In the present paper, the senses are further classified into some categories.

Keywords: cognitive, contexts, semantics, senses

Procedia PDF Downloads 199

Authors: Daliyah S. Aljutaili, Redna A. Almutlaq, Suha A. Alharbi, Dina M. Ibrahim

Abstract:

All currencies around the world look very different from each other. For instance, the size, color, and pattern of the paper are different. With the development of modern banking services, automatic methods for paper currency recognition become important in many applications like vending machines. One of the currency recognition architecture’s phases is Feature detection and description. There are many algorithms that are used for this phase, but they still have some disadvantages. This paper proposes a feature detection algorithm, which merges the advantages given in the current SIFT and SURF algorithms, which we call, Speeded up Robust Scale-Invariant Feature Transform (SR-SIFT) algorithm. Our proposed SR-SIFT algorithm overcomes the problems of both the SIFT and SURF algorithms. The proposed algorithm aims to speed up the SIFT feature detection algorithm and keep it robust. Simulation results demonstrate that the proposed SR-SIFT algorithm decreases the average response time, especially in small and minimum number of best key points, increases the distribution of the number of best key points on the surface of the currency. Furthermore, the proposed algorithm increases the accuracy of the true best point distribution inside the currency edge than the other two algorithms.

Keywords: currency recognition, feature detection and description, SIFT algorithm, SURF algorithm, speeded up and robust features

Procedia PDF Downloads 216

Authors: Saeedeh Barzegar Khalilsaraei, Manoocheher Kelarestaghi, Farshad Eshghi

Abstract:

The vehicle color, as a prominent and stable feature, helps to identify a vehicle more accurately. As a result, vehicle color recognition is of great importance in intelligent transportation systems. Unlike conventional methods which use only a single Convolutional Neural Network (CNN) for feature extraction or classification, in this paper, four CNNs, with different architectures well-performing in different classes, are trained to extract various features from the input image. To take advantage of the distinct capability of each network, the multiple outputs are combined using a stack generalization algorithm as an ensemble technique. As a result, the final model performs better than each CNN individually in vehicle color identification. The evaluation results in terms of overall average accuracy and accuracy variance show the proposed method’s outperformance compared to the state-of-the-art rivals.

Keywords: Vehicle Color Recognition, Ensemble Algorithm, Stack Generalization, Convolutional Neural Network

Procedia PDF Downloads 59

Authors: Jayashree Aanand, Gajjam

Abstract:

The grammatical tradition in Sanskrit literature emphasizes the importance of the correct use of Sanskrit words or linguistic units (sādhu śabda) that brings the meritorious values, denying the attribution of the same religious merit to the incorrect use of Sanskrit words (asādhu śabda) or the vernacular or corrupt forms (apa-śabda or apabhraṁśa), even though they may help in communication. The current research, the culmination of the doctoral research on sentence definition, studies the difference among the comprehension of both correct and incorrect word forms in Sanskrit and Marathi languages in India. Based on the total of 19 experiments (both web-based and classroom-controlled) on approximately 900 Indian readers, it is found that while the incorrect forms in Sanskrit are comprehended with lesser accuracy than the correct word forms, no such difference can be seen for the Marathi language. It is interpreted that the incorrect word forms in the native language or in the language which is spoken daily (such as Marathi) will pose a lesser cognitive load as compared to the language that is not spoken on a daily basis but only used for reading (such as Sanskrit). The theoretical base for the research problem is as follows: among the three main schools of Language Science in ancient India, the Vaiyākaraṇas (Grammarians) hold that the corrupt word forms do have their own expressive power since they convey meaning, while as the Mimāṁsakas (the Exegesists) and the Naiyāyikas (the Logicians) believe that the corrupt forms can only convey the meaning indirectly, by recalling their association and similarity with the correct forms. The grammarians argue that the vernaculars that are born of the speaker’s inability to speak proper Sanskrit are regarded as degenerate versions or fallen forms of the ‘divine’ Sanskrit language and speakers who could not use proper Sanskrit or the standard language were considered as Śiṣṭa (‘elite’). The different ideas of different schools strictly adhere to their textual dispositions. For the last few years, sociolinguists have agreed that no variety of language is inherently better than any other; they are all the same as long as they serve the need of people that use them. Although the standard form of a language may offer the speakers some advantages, the non-standard variety is considered the most natural style of speaking. This is visible in the results. If the incorrect word forms incur the recall of the correct word forms in the reader as the theory suggests, it would have added one extra step in the process of sentential cognition leading to more cognitive load and less accuracy. This has not been the case for the Marathi language. Although speaking and listening to the vernaculars is the common practice and reading the vernacular is not, Marathi readers have readily and accurately comprehended the incorrect word forms in the sentences, as against the Sanskrit readers. The primary reason being Sanskrit is spoken and also read in the standard form only and the vernacular forms in Sanskrit are not found in the conversational data.

Keywords: experimental sociolinguistics, grammaticality and meaningfulness, Marathi, Sanskrit

Procedia PDF Downloads 107

Authors: Nawab Yousuf Ali, S. Golam, A. Ameer, Ashok Toru Roy

Abstract:

This paper develops new morphological rules suitable for Bangla repetition words to be incorporated into an inter lingua representation called Universal Networking Language (UNL). The proposed rules are to be used to combine verb roots and their inflexions to produce words which are then combined with other similar types of words to generate repetition words. This paper outlines the format of morphological rules for different types of repetition words that come from verb roots based on the framework of UNL provided by the UNL centre of the Universal Networking Digital Language (UNDL) foundation.

Keywords: Universal Networking Language (UNL), universal word (UW), head word (HW), Bangla-UNL Dictionary, morphological rule, enconverter (EnCo)

Procedia PDF Downloads 291

Authors: Lanchi Xie, Zhihui Li, Zhigang Li, Guiqiang Wang, Lei Xu, Yuwen Yan

Abstract:

Image-based facial features can be classified into category recognition features and individual recognition features. Current automated face recognition systems extract a specific feature vector of different dimensions from a facial image according to their pre-trained neural network. However, to improve the efficiency of parameter calculation, an algorithm generally reduces the image details by pooling. The operation will overlook the details concerned much by forensic experts. In our experiment, we adopted a variety of face recognition algorithms based on deep learning, compared a large number of naturally collected face images with the known data of the same person's frontal ID photos. Downscaling and manual handling were performed on the testing images. The results supported that the facial recognition algorithms based on deep learning detected structural and morphological information and rarely focused on specific markers such as stains and moles. Overall performance, distribution of genuine scores and impostor scores, and likelihood ratios were tested to evaluate the accuracy of biometric systems and forensic experts. Experiments showed that the biometric systems were skilled in distinguishing category features, and forensic experts were better at discovering the individual features of human faces. In the proposed approach, a fusion was performed at the score level. At the specified false accept rate, the framework achieved a lower false reject rate. This paper contributes to improving the interpretability of the objective method of facial comparison and provides a novel method for human-machine collaboration in this field.

Keywords: likelihood ratio, automated facial recognition, facial comparison, biometrics

Procedia PDF Downloads 111

Authors: Jun Gil Ahn, Jong Kang Park, Jong Tae Kim

Abstract:

In this paper, we analyze major components of activity recognition (AR) in wearable device with 9-axis sensors and sensor fusion filters. 9-axis sensors commonly include 3-axis accelerometer, 3-axis gyroscope and 3-axis magnetometer. We chose sensor fusion filters as Kalman filter and Direction Cosine Matrix (DCM) filter. We also construct sensor fusion data from each activity sensor data and perform classification by accuracy of AR using Naïve Bayes and SVM. According to the classification results, we observed that the DCM filter and the specific combination of the sensing axes are more effective for AR in wearable devices while classifying walking, running, ascending and descending.

Keywords: accelerometer, activity recognition, directiona cosine matrix filter, gyroscope, Kalman filter, magnetometer

Procedia PDF Downloads 310

Authors: Koray U. Erbas

Abstract:

Neural networks are appealing for many applications since they are able to learn complex non-linear relationships between input and output data. As the number of neurons and layers in a neural network increase, it is possible to represent more complex relationships with automatically extracted features. Nowadays Deep Neural Networks (DNNs) are widely used in Computer Vision problems such as; classification, object detection, segmentation image editing etc. In this work, Facial Emotion Recognition task is performed by proposed Convolutional Neural Network (CNN)-based DNN architecture using FER2013 Dataset. Moreover, the effects of different hyperparameters (activation function, kernel size, initializer, batch size and network size) are investigated and ablation study results for Pooling Layer, Dropout and Batch Normalization are presented.

Keywords: convolutional neural network, deep learning, deep learning based FER, facial emotion recognition

Procedia PDF Downloads 235

Authors: Carlos Vera, Tetyana Baydyk, Ernst Kussul, Graciela Velasco, Miguel Aparicio

Abstract:

This article describes the Random Subspace Neural Classifier (RSC) for the recognition of meteors in the night sky. We used images of meteors entering the atmosphere at night between 8:00 p.m.-5: 00 a.m. The objective of this project is to classify meteor and star images (with stars as the image background). The monitoring of the sky and the classification of meteors are made for future applications by scientists. The image database was collected from different websites. We worked with RGB-type images with dimensions of 220x220 pixels stored in the BitMap Protocol (BMP) format. Subsequent window scanning and processing were carried out for each image. The scan window where the characteristics were extracted had the size of 20x20 pixels with a scanning step size of 10 pixels. Brightness, contrast and contour orientation histograms were used as inputs for the RSC. The RSC worked with two classes and classified into: 1) with meteors and 2) without meteors. Different tests were carried out by varying the number of training cycles and the number of images for training and recognition. The percentage error for the neural classifier was calculated. The results show a good RSC classifier response with 89% correct recognition. The results of these experiments are presented and discussed.

Keywords: contour orientation histogram, meteors, night sky, RSC neural classifier, stars

Procedia PDF Downloads 122

Authors: Wanqing You, Kai Qian, Xi He, Ying Qian

Abstract:

In this paper, the potential security issues brought by the virtualization of a Software Defined Networks (SDN) would be analyzed. The virtualization of SDN is achieved by FlowVisor (FV). With FV, a physical network is divided into multiple isolated logical networks while the underlying resources are still shared by different slices (isolated logical networks). However, along with the benefits brought by network virtualization, it also presents some issues regarding security. By examining security issues existing in an OpenFlow network, which uses FlowVisor to slice it into multiple virtual networks, we hope we can get some significant results and also can get further discussions among the security of SDN virtualization.

Keywords: SDN, network, virtualization, security

Procedia PDF Downloads 398

Authors: Paula Buldres, Jorge Toledo

Abstract:

Objectives: This study aimed to characterize potential probiotic strains isolated from canine breast milk for use in dogs with enteropathies. Methodology: Six canine breast milk strains, one canine colostrum strain, and one control porcine breast milk strain were characterized. According to its functional properties of resistance to acids, different concentrations of bile salts, and pancreatin, its presumptive properties of safety and inhibitory effect on pathogens, non-cytotoxic characteristics, and adhesion to the intestine. The immunomodulatory effect of formulations with better probiotic characterization in vitro and in vivo was also analyzed. Results: Two strains characterized as potential probiotics were obtained, which corresponded to the canine strains (TUCO-16 and TUCO-17), presenting resistance to acidic pH, bile salts, and pancreatin, as well as an inhibitory effect on pathogenic Escherichia coli, Salmonella sp., and Clostridium perfringens. Strains TUCO-16 and TUCO-17 induced a significant increase in the expression of TNF-α and IL-8 in canine macrophages, respectively. Expression analyses of pattern recognition receptors in DH82 cells suggest that TUCO-16 and TUCO-17 might increase the TLR2 expression marker, and porcine strain (TUCO-4) increases the NOD2 expression marker. Based on the count obtained and the encapsulation yield, the best formulations correspond to FOS-Inulin for the TUCO-17 and TUCO-4 strains; Maltodextrin-Inulin for TUCO-16. All the strains are non-cytotoxic. The strain that showed the highest adhesion to intestinal epithelial cells was TUCO-17 with the FOS-Inulin formulation. On the other hand, the probiotics decreased the expression of pro-inflammatory markers in vivo, both in the intestine and in the spleen of mice. Conclusion: The combination of these three strains under study (TUCO-16, TUCO-17, and TUCO-4) would cover the probiotic properties in formulation and immunomodulation of all the markers under study.

Keywords: probiotics, gastrointestinal infec, dog, probiotic formulation, immunomodulatory probiotics

Procedia PDF Downloads 45

Authors: Ataa A. Said, Elsayed A. Abuotabl, Gehan F. Abdel Raoof, Khaled Y. Mohamed

Abstract:

The aim of this study is to evaluate antihyperglycaemic and antihyperlipidemic activities of Pleiogynium timorense (DC.) Leenh (Anacardiaceae) seeds as well as to isolate and identify the bioactive compounds. Antihyperglycaemic effect was evaluated by measuring the effect of two dose levels (150 and 300 mg/kg) of 70% methanol extract of Pleiogynium timorense seeds on blood glucose level when administered 45 minutes before glucose loading. In addition, the effect of the plant extract on the lipid profile was determined by measuring serum total lipids (TL), total cholesterol (TC), triglycerides (TG), high density lipoprotein cholesterol (HDL-C) and low density lipoprotein cholesterol (LDL-C). Furthermore, the bioactive compounds were isolated and identified by chromatographic and spectrometric methods.The results showed that the methanolic extract of the seeds significantly reduced the levels of blood glucose,(TL), (TC), (TG) and (LDL-C) but no significant effect on (HDL-C) comparing with control group. Furthermore, four phenolic compound were isolated which were identified as; catechin, gallic acid, para methoxy benzaldehyde and pyrogallol which were isolated for the first time from the plant. In addition sulphur -containing compound (sulpholane) was isolated for the first time from the plant and from the family. To our knowledge, this is the first study about antihyperglycaemicand antihyperlipidemic activities of the seeds of Pleiogyniumtimorense and its bioactive compounds. So, the methanolic extract of the seeds of Pleiogynium timorense could be a step towards the development of new antihyperglycaemic and antihyperlipidemic drugs.

Keywords: antihyperglycaemic, bioactive compounds, phenolic, Pleiogynium timorense, seeds

Procedia PDF Downloads 199

Authors: R. Gounina-Allouane, I. Moussaoui, N. Boukahel

Abstract:

Bacillus antimicrobial metabolites, especially those of Bacillus thuringiensis (Bt), are of great interest for research because of health risks generated by the excessive use of chemical additives as well as the propagation of resistant microbial strains, caused by the massive treatment with antibiotics. The objective of this study was the selection of Bt strains producing antimicrobial peptides (bacteriocins), and the partial purification of the most powerful bacteriocins, then the determination of their spectra of antimicrobial action. A collection of twenty one Bt strains isolated from soil at Boumerdès (northern Algeria) was used for screening strains having an antagonistic activity against phylogenetically closed bacteria. Spectra of antagonistic activity of two selected strains was determined against other Bt strains, Gram positive and Gram negative bacterial strains of clinical origin and others from ATCC collection as well as yeasts isolated in human dermatology. Bacteriocins of these two strains were partially purified and their effect on the kinetics of growth of the most sensitive microbial strains was studied. The bacteriocinogenic strains were biochemically characterized and their sensitivity to antibiotics was studied.

Keywords: antimicrobial peptides, Bacillus thuringiensis, bacteriocin, partial purification

Procedia PDF Downloads 413

Authors: Wanhyun Cho, Soonja Kang, Sangkyoon Kim, Soonyoung Park

Abstract:

In this paper, we present the human action recognition method using the variational Bayesian HMM with the Dirichlet process mixture (DPM) of the Gaussian-Wishart emission model (GWEM). First, we define the Bayesian HMM based on the Dirichlet process, which allows an infinite number of Gaussian-Wishart components to support continuous emission observations. Second, we have considered an efficient variational Bayesian inference method that can be applied to drive the posterior distribution of hidden variables and model parameters for the proposed model based on training data. And then we have derived the predictive distribution that may be used to classify new action. Third, the paper proposes a process of extracting appropriate spatial-temporal feature vectors that can be used to recognize a wide range of human behaviors from input video image. Finally, we have conducted experiments that can evaluate the performance of the proposed method. The experimental results show that the method presented is more efficient with human action recognition than existing methods.

Keywords: human action recognition, Bayesian HMM, Dirichlet process mixture model, Gaussian-Wishart emission model, Variational Bayesian inference, prior distribution and approximate posterior distribution, KTH dataset

Procedia PDF Downloads 329

Authors: B. R. Campomanes-Alvarez, P. Quiros, B. Fernandez

Abstract:

Automatic Speech Recognition (ASR) is a machine-based process of decoding and transcribing oral speech. A typical ASR system receives acoustic input from a speaker or an audio file, analyzes it using algorithms, and produces an output in the form of a text. Some speech recognition systems use Hidden Markov Models (HMMs) to deal with the temporal variability of speech and Gaussian Mixture Models (GMMs) to determine how well each state of each HMM fits a short window of frames of coefficients that represents the acoustic input. Another way to evaluate the fit is to use a feed-forward neural network that takes several frames of coefficients as input and produces posterior probabilities over HMM states as output. Deep neural networks (DNNs) that have many hidden layers and are trained using new methods have been shown to outperform GMMs on a variety of speech recognition systems. Acoustic models for state-of-the-art ASR systems are usually training on massive amounts of data. However, audio files with their corresponding transcriptions can be difficult to obtain, especially in the Spanish language. Hence, in the case of these low-resource scenarios, building an ASR model is considered as a complex task due to the lack of labeled data, resulting in an under-trained system. Semi-supervised learning approaches arise as necessary tasks given the high cost of transcribing audio data. The main goal of this proposal is to develop a procedure based on acoustic semi-supervised learning for Spanish ASR systems by using DNNs. This semi-supervised learning approach consists of: (a) Training a seed ASR model with a DNN using a set of audios and their respective transcriptions. A DNN with a one-hidden-layer network was initialized; increasing the number of hidden layers in training, to a five. A refinement, which consisted of the weight matrix plus bias term and a Stochastic Gradient Descent (SGD) training were also performed. The objective function was the cross-entropy criterion. (b) Decoding/testing a set of unlabeled data with the obtained seed model. (c) Selecting a suitable subset of the validated data to retrain the seed model, thereby improving its performance on the target test set. To choose the most precise transcriptions, three confidence scores or metrics, regarding the lattice concept (based on the graph cost, the acoustic cost and a combination of both), was performed as selection technique. The performance of the ASR system will be calculated by means of the Word Error Rate (WER). The test dataset was renewed in order to extract the new transcriptions added to the training dataset. Some experiments were carried out in order to select the best ASR results. A comparison between a GMM-based model without retraining and the DNN proposed system was also made under the same conditions. Results showed that the semi-supervised ASR-model based on DNNs outperformed the GMM-model, in terms of WER, in all tested cases. The best result obtained an improvement of 6% relative WER. Hence, these promising results suggest that the proposed technique could be suitable for building ASR models in low-resource environments.

Keywords: automatic speech recognition, deep neural networks, machine learning, semi-supervised learning

Procedia PDF Downloads 322

Authors: Nurcihan Hacioglu, Cigdem Gul, Murat Tosunoglu

Abstract:

In this article, the antibiogram and heavy metal resistance profile of the bacteria isolated from total 34 studied animals (Pelophylax ridibundus = 12, Mauremys rivulata = 14, Natrix natrix = 8) captured around the Biga Stream, are described. There was no database information on antibiogram and heavy metal resistance profile of bacteria from these area’s amphibians and reptiles. In this study, a total of 200 bacteria were successfully isolated from cloaca and oral samples of the aquatic amphibians and reptiles as well as from the water sample. According to Jaccard’s similarity index, the degree of similarity in the bacterial flora was quite high among the amphibian and reptile species under examination, whereas it was different from the bacterial diversity in the water sample. The most frequent isolates were A. hydrophila (31.5%), B. pseudomallei (8.5%), and C. freundii (7%). The total numbers of bacteria obtained were as follows: 45 in P. ridibundus, 45 in N. natrix 30 in M. rivulata, and 80 in the water sample. The result showed that cefmetazole was the most effective antibiotic to control the bacteria isolated in this study and that approximately 93.33% of the bacterial isolates were sensitive to this antibiotic. The Multiple Antibiotic Resistances (MAR) index indicated that P. ridibundus (0.95) > N. natrix (0.89) > M. rivulata (0.39). Furthermore, all the tested heavy metals (Pb+2, Cu+2, Cr+3, and Mn+2) inhibit the growth of the bacterial isolates at different rates. Therefore, it indicated that the water source of the animals was contaminated with both antibiotic residues and heavy metals.

Keywords: bacteriological quality, amphibian, reptile, antibiotic, heavy metal resistance

Procedia PDF Downloads 362

Authors: Hamza Iftikhar

Abstract:

The research study focuses on the transgender community's gender recognition challenges. It is one of the issues for the transgender community, interacting directly with the difficulties of gender identity and the lives of these people who are facing gender disapproval from society. This study investigates the major flaws of the transgender act. The study's goal is to look into the strange implications of self-perceived identity as a policy tool for transgender recognition. This policy tool jeopardises the rights of Pakistan's indigenous gender-variant people as well as the country's legal and social framework. Qualitative research using semi structured interviews will be carried out. This study proposes developing a scheme for mainstreaming gender-variant people on the basis of the Pakistani Constitution, Supreme Court guidelines, and internationally recognised principles of law. This would necessitate a thorough review of current law using a new approach and reference point.

Keywords: transgender act, self perceived identity, gender variant, policy tool

Procedia PDF Downloads 100

Authors: Yuh-Jen Wu, Chun-Min Lin

Abstract:

Homophonic words are common in Mandarin Chinese which belongs to the tonal language family. Using homophonic cues to study foreign languages is one of the learning techniques of mnemonics that can aid the retention and retrieval of information in the human memory. When learning difficult foreign words, some learners transpose them with words in a language they are familiar with to build an association and strengthen working memory. These phonological clues are beneficial means for novice language learners. In the classroom, if mnemonic skills are used at the appropriate time in the instructional sequence, it may achieve their maximum effectiveness. For Chinese-speaking students, proper use of Chinese-English homophonic word pairs may help them learn difficult vocabulary. In this study, a database program is developed by employing Visual Basic. The database contains two corpora, one with Chinese lexical items and the other with English ones. The Chinese corpus contains 59,053 Chinese words that were collected by a web crawler. The pronunciations of this group of words are compared with words in an English corpus based on WordNet, a lexical database for the English language. Words in both databases with similar pronunciation chunks and batches are detected. A total of approximately 1,000 Chinese lexical items are located in the preliminary comparison. These homophonic word pairs can serve as a valuable tool to assist Chinese-speaking students in learning and memorizing new English vocabulary.

Keywords: Chinese, corpus, English, homophonic words, vocabulary

Procedia PDF Downloads 158

Authors: Elena E. Lyakso, Olga V. Frolova, Yuri N. Matveev, Aleksey S. Grigorev, Alexander S. Nikolaev, Viktor A. Gorodnyi

Abstract:

The study of the children’s emotional sphere depending on age and psychoneurological state is of great importance for the design of educational programs for children and their social adaptation. Atypical development may be accompanied by violations or specificities of the emotional sphere. To study characteristics of the emotional state reflection in the voice and speech features of children, the perceptual study with the participation of adults and the automatic recognition of speech were conducted. Speech of children with typical development (TD), with Down syndrome (DS), and with autism spectrum disorders (ASD) aged 6-12 years was recorded. To obtain emotional speech in children, model situations were created, including a dialogue between the child and the experimenter containing questions that can cause various emotional states in the child and playing with a standard set of toys. The questions and toys were selected, taking into account the child’s age, developmental characteristics, and speech skills. For the perceptual experiment by adults, test sequences containing speech material of 30 children: TD, DS, and ASD were created. The listeners were 100 adults (age 19.3 ± 2.3 years). The listeners were tasked with determining the children’s emotional state as “comfort – neutral – discomfort” while listening to the test material. Spectrographic analysis of speech signals was conducted. For automatic recognition of the emotional state, 6594 speech files containing speech material of children were prepared. Automatic recognition of three states, “comfort – neutral – discomfort,” was performed using automatically extracted from the set of acoustic features - the Geneva Minimalistic Acoustic Parameter Set (GeMAPS) and the extended Geneva Minimalistic Acoustic Parameter Set (eGeMAPS). The results showed that the emotional state is worse determined by the speech of TD children (comfort – 58% of correct answers, discomfort – 56%). Listeners better recognized discomfort in children with ASD and DS (78% of answers) than comfort (70% and 67%, respectively, for children with DS and ASD). The neutral state is better recognized by the speech of children with ASD (67%) than by the speech of children with DS (52%) and TD children (54%). According to the automatic recognition data using the acoustic feature set GeMAPSv01b, the accuracy of automatic recognition of emotional states for children with ASD is 0.687; children with DS – 0.725; TD children – 0.641. When using the acoustic feature set eGeMAPSv01b, the accuracy of automatic recognition of emotional states for children with ASD is 0.671; children with DS – 0.717; TD children – 0.631. The use of different models showed similar results, with better recognition of emotional states by the speech of children with DS than by the speech of children with ASD. The state of comfort is automatically determined better by the speech of TD children (precision – 0.546) and children with ASD (0.523), discomfort – children with DS (0.504). The data on the specificities of recognition by adults of the children’s emotional state by their speech may be used in recruitment for working with children with atypical development. Automatic recognition data can be used to create alternative communication systems and automatic human-computer interfaces for social-emotional learning. Acknowledgment: This work was financially supported by the Russian Science Foundation (project 18-18-00063).

Keywords: autism spectrum disorders, automatic recognition of speech, child’s emotional speech, Down syndrome, perceptual experiment

Procedia PDF Downloads 169

Authors: K. Veluraja

Abstract:

Glycan of glycolipids and glycoproteins is playing a significant role in living systems particularly in molecular recognition processes. Molecular recognition processes are attributed to their occurrence on the surface of the cell, sequential arrangement and type of sugar molecules present in the oligosaccharide structure and glyosidic linkage diversity (glycoinformatics) and conformational diversity (glycoconformatics). Molecular Dynamics Simulation study is a theoretical-cum-computational tool successfully utilized to establish glycoconformatics of glycan. The study on various oligosaccharides of glycan clearly indicates that oligosaccharides do exist in multiple conformational states and these conformational states arise due to the flexibility associated with a glycosidic torsional angle (φ,ψ) . As an example: a single disaccharide structure NeuNacα(2-3) Gal exists in three different conformational states due to the differences in the preferential value of glycosidic torsional angles (φ,ψ). Hence establishing three dimensional structural and conformational models for glycan (cartesian coordinates of every individual atoms of an oligosaccharide structure in a preferred conformation) is quite crucial to understand various molecular recognition processes such as glycan-toxin interaction and glycan-virus interaction. The gycoconformatics models obtained for various glycan through Molecular Dynamics Simulation stored in our 3DSDSCAR (3DSDSCAR.ORG) a public domain database and its utility value in understanding the molecular recognition processes and in drug design venture will be discussed.

Keywords: glycan, glycoconformatics, molecular dynamics simulation, oligosaccharide

Procedia PDF Downloads 115

Authors: M. C. Padma, R. J. Prathibha

Abstract:

Parts of speech tagging is the process of assigning appropriate parts of speech tags to the words in a given text. The critical or crucial information needed for tagging a word come from its internal structure rather from its neighboring words. The internal structure of a word comprises of its morphological features and grammatical information. This paper presents a morpheme based parts of speech tagger for Kannada language. This proposed work uses hierarchical tag set for assigning tags. The system is tested on some Kannada words taken from EMILLE corpus. Experimental result shows that the performance of the proposed system is above 90%.

Keywords: hierarchical tag set, morphological analyzer, natural language processing, paradigms, parts of speech

Procedia PDF Downloads 270

Authors: Panagiotis Pentaris

Abstract:

The issue of social and legal recognition of LGBTQ families is of high importance when exploring the possibility of a family. Of equal importance is the fact that both society and the individual contribute to the overall recognition of LGBTQ families. This paper is a conceptual discussion, by methodology, of both sides; it uses a method of constructive analysis to expound on this issue. This method’s aim is to broaden conceptual theory, and introduce a new relationship between concepts that were previously not associated by evidence. This exploration has found that LGBTQ realities from an international perspective may differ and both legal and social rights are critical toward self-consciousness and the formation of a family. This paper asserts that internalised and historic oppression of LGBTQ individuals, places them, not always and not in all places, in a disadvantageous position as far as engaging with the potential of forming a family goes. The paper concludes that lack of social recognition and internalised oppression are key barriers regarding LGBTQ families.

Keywords: family, gay, self-worth, LGBTQ, social rights

Procedia PDF Downloads 112

Authors: R. Gounina-Allouane, I. Moussaoui, N. Boukahel

Abstract:

Bacillus antimicrobial metabolites, especially those of Bacillus thuringiensis (Bt), are of great interest for research because of health risks generated by the excessive use of chemical additives as well as the propagation of resistant microbial strains, caused by the massive treatment with antibiotics. The objective of this study was the selection of Bt strains producing antimicrobial peptides (bacteriocins), and the partial purification of the most powerful bacteriocins, then the determination of their spectra of antimicrobial action. A collection of twenty one Bt strains isolated from soil at Boumerdès (northern of Algeria) was used for screening strains having an antagonistic activity against phylogenetically closed bacteria. Spectra of antagonistic activity of two selected strains was determined against other Bt strains, Gram positive and Gram negative bacterial strains of clinical origin and others from ATCC collection as well as yeasts isolated in human dermatology. Bacteriocins of these two strains were partially purified and their effect on the kinetics of growth of the most sensitive microbial strains was studied. The bacteriocinogenic strains were biochemically characterized and their sensitivity to antibiotics was studied.

Keywords: antimicrobial peptides, Bacillus thuringiensis, bacteriocin, partial purification

Procedia PDF Downloads 338

Authors: Harikrishna Yadav Nanganuru, Narasimhulu Korrapati

Abstract:

Over the past many years, many sites in the world have been contaminated with heavy metals, which are the largest class of contaminants. Lead is one of the toxic heavy metals contaminated in the environment. Lead is not biodegradable, that’s why it is accumulated in the human body and impacts all the systems of the human body when it has been taken by humans. The accumulation of lead in the water environment has been showing adverse effects on the public health. So the removal of lead from the water environment by the biosorption process, which is emerged as a potential method for the lead removal, is an efficient approach. This work was focused to examine the removal of Lead [Pb (II)] ions from aqueous solution and effluent from battery industry. Lead contamination in water is a widespread problem throughout the world and mainly results from lead acid battery manufacturing effluent. In this work, isolated bacteria from wastewater of lead acid battery industry has been utilized for the removal of lead. First effluent from the lead acid battery industry was characterized by the inductively coupled plasma atomic emission spectrometry (ICP – AES). Then the bacteria was isolated from the effluent and used it’s immobilized dead mass for the biosorption of lead. Scanning electron microscopic (SEM) and Atomic force microscopy (AFM) studies clearly suggested that the Lead (Pb) was adsorbed efficiently. The adsorbed percentage of lead (II) from waste was 97.40 the concentration of lead (II) is measured by Atomic Absorption Spectroscopy (AAS). From the result of AAS it can be concluded that immobilized isolated dead mass was well efficient and useful for biosorption of lead contaminated waste water.

Keywords: biosorption, ICP-AES, lead (Pb), SEM

Procedia PDF Downloads 359