Search results for: speech to text

Authors: Rosa L. Figueroa, Christopher A. Flores, Qing Zeng-Treitler

Abstract:

A regular expression consists of sequence characters which allow describing a text path. Usually, in clinical research, regular expressions are manually created by programmers together with domain experts. Lately, there have been several efforts to investigate how to generate them automatically. This article presents a text classification algorithm based on regexes. The algorithm named REX was designed, and then, implemented as a simplified method to create regexes to classify Spanish text automatically. In order to classify ambiguous cases, such as, when multiple labels are assigned to a testing example, REX includes an information gain method Two sets of data were used to evaluate the algorithm’s effectiveness in clinical text classification tasks. The results indicate that the regular expression based classifier proposed in this work performs statically better regarding accuracy and F-measure than Support Vector Machine and Naïve Bayes for both datasets.

Keywords: information gain, regular expressions, smith-waterman algorithm, text classification

Procedia PDF Downloads 289

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 158

Authors: Indira Kothalanka

Abstract:

For individuals born with a cleft palate (CP), there is no separation between the nasal cavity and the oral cavity, due to which they cannot build up enough air pressure in the mouth for speech. Therefore, it is common for them to have speech problems. Common cleft type speech errors include abnormal articulation (compensatory or obligatory) and abnormal resonance (hyper, hypo and mixed nasality). These are generally resolved after palate repair. However, in some individuals, articulation problems do persist even after the palate repair. Such individuals develop variant articulations in an attempt to compensate for the inability to produce the target phonemes. A spectrographic analysis is used to investigate the compensatory articulatory behaviours of pressure consonants in the speech of 10 Telugu speaking individuals aged between 7-17 years with a history of cleft palate. Telugu is a Dravidian language which is spoken in Andhra Pradesh and Telangana states in India. It is a language with the third largest number of native speakers in India and the most spoken Dravidian language. The speech of the informants is analysed using single word list, sentences, passage and conversation. Spectrographic analysis is carried out using PRAAT, speech analysis software. The place and manner of articulation of consonant sounds is studied through spectrograms with the help of various acoustic cues. The types of compensatory articulation identified are glottal stops, palatal stops, uvular, velar stops and nasal fricatives which are non-native in Telugu.

Keywords: cleft palate, compensatory articulation, spectrographic analysis, PRAAT

Procedia PDF Downloads 416

Authors: P. S. Jagadeesh Kumar, S. Meenakshi Sundaram, Wenli Hu, Yang Yung

Abstract:

In 3D video games, the dominance of production is unceasingly growing with a protruding level of affordability in terms of budget. Afterward, the automation of speech-lip synchronization technique is customarily onerous and has advanced a critical research subject in virtual reality based 3D video games. This paper presents one of these automatic tools, precisely riveted on the synchronization of the speech and the lip movement of the game characters. A robust and precise speech recognition segment that systematized with Algebraic Code Excited Linear Prediction method is developed which unconventionally delivers lip sync results. The Algebraic Code Excited Linear Prediction algorithm is constructed on that used in code-excited linear prediction, but Algebraic Code Excited Linear Prediction codebooks have an explicit algebraic structure levied upon them. This affords a quicker substitute to the software enactments of lip sync algorithms and thus advances the superiority of service factors abridged production cost.

Keywords: algebraic code excited linear prediction, speech-lip synchronization, video games, virtual reality

Procedia PDF Downloads 439

Authors: D. Beziakina, E. Bulgakova

Abstract:

This study continues the research on Tuvinian intonation and presents a general cross-dialect analysis of intonation of Tuvinian declarative utterances, specifically the character of the tone movement in order to test the hypothesis about the prevalence of level tone in some Tuvinian dialects. The results of the analysis of basic pitch characteristics of Tuvinian speech (in general and in comparison with two other Turkic languages - Uzbek and Azerbaijani) are also given in this paper. The goal of our work was to obtain the ranges of pitch parameter values typical for Tuvinian speech. Such language-specific values can be used in speaker identification systems in order to get more accurate results of ethnic speech analysis. We also present the results of a cross-dialect analysis of declarative intonation in the poorly studied Tuvinian language.

Keywords: speech analysis, statistical analysis, speaker recognition, identification of person

Procedia PDF Downloads 439

Authors: Maisa Haj-Tas, Jehad Alaraifi

Abstract:

The significance of the study: This retrospective study examined the speech and language profiles of patients who received clinical services at the University of Jordan Hearing and Speech Clinic (UJ-HSC) from 2009 to 2014. The UJ-HSC clinic is located in the capital Amman and was established in the late 1990s. It is the first hearing and speech clinic in Jordan and one of first speech and hearing clinics in the Middle East. This clinic provides services to an annual average of 2000 patients who are diagnosed with different communication disorders. Examining the speech and language profiles of patients in this clinic could provide an insight about the most common disorders seen in patients who attend similar clinics in Jordan. It could also provide information about community awareness of the role of speech therapists in the management of speech and language disorders. Methodology: The researchers examined the clinical records of 1140 patients (797 males and 343 females) who received clinical services at the UJ-HSC between the years 2009 and 2014 for the purpose of data analysis for this study. The main variables examined in the study were disorder type and gender. Participants were divided into four age groups: children, adolescents, adults, and older adults. The examined disorders were classified as either speech disorders, language disorders, or dysphagia (i.e., swallowing problems). The disorders were further classified as childhood language impairments, articulation disorders, stuttering, cluttering, voice disorders, aphasia, and dysphagia. Results: The results indicated that the prevalence for language disorders was the highest (50.7%) followed by speech disorders (48.3%), and dysphagia (0.9%). The majority of patients who were seen at the JU-HSC were diagnosed with childhood language impairments (47.3%) followed consecutively by articulation disorders (21.1%), stuttering (16.3%), voice disorders (12.1%), aphasia (2.2%), dysphagia (0.9%), and cluttering (0.2%). As for gender, the majority of patients seen at the clinic were males in all disorders except for voice disorders and cluttering. Discussion: The results of the present study indicate that the majority of examined patients were diagnosed with childhood language impairments. Based on this result, the researchers suggest that there seems to be a high prevalence of childhood language impairments among children in Jordan compared to other types of speech and language disorders. The researchers also suggest that there is a need for further examination of the actual prevalence data on speech and language disorders in Jordan. The fact that many of the children seen at the UJ-HSC were brought to the clinic either as a result of parental concern or teacher referral indicates that there seems to an increased awareness among parents and teachers about the services speech pathologists can provide about assessment and treatment of childhood speech and language disorders. The small percentage of other disorders (i.e., stuttering, cluttering, dysphasia, aphasia, and voice disorders) seen at the UJ-HSC may indicate a little awareness by the local community about the role of speech pathologists in the assessment and treatment of these disorders.

Keywords: clinic, disorders, language, profile, speech

Procedia PDF Downloads 287

Authors: Syed Basit Ali Shah Bukhari, Yan Qiang, Saad Abdul Rauf, Syed Saqlaina Bukhari

Abstract:

Text Classification is the methodology to classify any given text into the respective category from a given set of categories. It is highly important and vital to use proper set of pre-processing , feature selection and classification techniques to achieve this purpose. In this paper we have used different ensemble techniques along with variance in feature selection parameters to see the change in overall accuracy of the result and also on some other individual class based features which include precision value of each individual category of the text. After subjecting our data through pre-processing and feature selection techniques , different individual classifiers were tested first and after that classifiers were combined to form ensembles to increase their accuracy. Later we also studied the impact of decreasing the classification categories on over all accuracy of data. Text classification is highly used in sentiment analysis on social media sites such as twitter for realizing people’s opinions about any cause or it is also used to analyze customer’s reviews about certain products or services. Opinion mining is a vital task in data mining and text categorization is a back-bone to opinion mining.

Keywords: Natural Language Processing, Ensemble Classifier, Bagging Classifier, AdaBoost

Procedia PDF Downloads 203

Authors: Jong Han Joo, Jung Hoon Lee, Young Sun Kim, Jae Young Kang, Seung Ho Choi

Abstract:

In this study, we propose a novel technique for acoustic echo suppression (AES) during speech recognition under barge-in conditions. Conventional AES methods based on spectral subtraction apply fixed weights to the estimated echo path transfer function (EPTF) at the current signal segment and to the EPTF estimated until the previous time interval. We propose a new approach that adaptively updates weight parameters in response to abrupt changes in the acoustic environment due to background noises or double-talk. Furthermore, we devised a voice activity detector and an initial time-delay estimator for barge-in speech recognition in communication networks. The initial time delay is estimated using log-spectral distance measure, as well as cross-correlation coefficients. The experimental results show that the developed techniques can be successfully applied in barge-in speech recognition systems.

Keywords: acoustic echo suppression, barge-in, speech recognition, echo path transfer function, initial delay estimator, voice activity detector

Procedia PDF Downloads 341

Authors: Khadija Rafi, Neha Jamil, Laiba Khalid, Meerub Nawaz, Mahwish Farooq

Abstract:

Speech articulation is a complex process to produce intelligible sounds with the help of precise movements of various structures within the vocal tract. All these structures in the vocal tract are named as articulators, which comprise lips, teeth, tongue, and palate. These articulators work together to produce a range of distinct phonemes, which happen to be the basis of language. It starts with the airstream from the lungs passing through the trachea and into oral and nasal cavities. When the air passes through the mouth, the tongue and the muscles around it form such coordination it creates certain sounds. It can be seen when the tongue is placed in different positions- sometimes near the alveolar ridge, soft palate, roof of the mouth or the back of the teeth which end up creating unique qualities of each phoneme. We can articulate vowels with open vocal tracts, but the height and position of the tongue is different every time depending upon each vowel, while consonants can be pronounced when we create obstructions in the airflow. For instance, the alphabet ‘b’ is a plosive and can be produced only by briefly closing the lips. Articulation disorders can not only affect communication but can also be a hurdle in speech production. To improve articulation skills for such individuals, doctors often recommend speech therapy, which involves various kinds of exercises like jaw exercises and tongue twisters. However, this disorder is more common in children who are going through developmental articulation issues right after birth, but in adults, it can be caused by injury, neurological conditions, or other speech-related disorders. In short, speech articulation is an essential aspect of productive communication, which also includes coordination of the specific articulators to produce different intelligible sounds, which are a vital part of spoken language.

Keywords: linguistics, speech articulation, speech therapy, language learning

Procedia PDF Downloads 30

Authors: Laurel Chikwado Madumere, Kevin O. Ugorji

Abstract:

A speech is said to be full of hate when it appropriates disparaging and vituperative locutions and/or appellations, which are riddled with prejudices and misconceptions about an antagonizing party on the grounds of gender, race, political orientation, religious affiliations, tribe, etc. Due largely to the dichotomies and polarities that exist in Nigeria across political ideological spectrum, tribal affiliations, and gender contradistinctions, there are possibilities for the existence of socioeconomic, religious and political conditions that would induce, provoke and catalyze hate speeches in Nigeria’s mainstream media. Therefore the aim of this paper is to investigate, using select daily newspapers in Nigeria, the extent and complexity of those likely hate speeches that emanate from the pluralism in Nigeria and to set in to relief, the discrepancies and contrariety in the interpretation of those hate words. To achieve the above, the paper shall be qualitative in orientation as it shall be using the Speech Act Theory of J. L. Austin and J. R. Searle to interpret and evaluate the hate speeches in the select Nigerian daily newspapers. Also this paper shall help to elucidate the conditions that generate hate, and inform the government and NGOs how best to approach those conditions and put an end to the possible violence and extremism that emanate from extreme cases of hate.

Keywords: extremism, gender, hate speech, pluralism, prejudice, speech act theory

Procedia PDF Downloads 118

Authors: Partha Protim Majumder, Sajeeb Das, Sharun Akter Khushbu

Abstract:

Bengali is widely spoken in the world, but Bengali speech recognition has not received much attention. Here, we are conducting the toughest task because it must be performed in a noisy place in our study. Another challenge we overcome is dealing with speeches and collecting data on third genders, and our approach is to recognize the gender in speeches. All of the Bangla speech samples used in this study were short and were taken from real-life situations. We employed the male, female, and third-gender categories of speech. In this study, we derive the feature from the spoken word. We used MFCC(1-20), ZCR,rolloff,spec_cen, RMSE, and chroma_stft. Here, we used the algorithms Gboost, Random Forest, K-Nearest Neighbors (KNN), Decision Tree, Naive Bayes, and Logistic Regression (LR) to assess the performance of recognition metrics, and we got the highest performance from random forest in recognizing the gender of the speeches.

Keywords: MFCC, ZCR, Bengali, LR, RMSE, roll-off, Gboost

Procedia PDF Downloads 30

Authors: Nor Diana Ahmad, Eric Atwell, Brandon Bennett

Abstract:

Text processing techniques for English have been developed for several decades. But for the Malay language, text processing methods are still far behind. Moreover, there are limited resources, tools for computational linguistic analysis available for the Malay language. Therefore, this research presents the use of natural language processing (NLP) in processing Malay translated Qur’an text. As the result, a new language resource for Malay translated Qur’an was created. This resource will help other researchers to build the necessary processing tools for the Malay language. This research also develops a simple question-answer prototype to demonstrate the use of the Malay Qur’an resource for text processing. This prototype has been developed using Python. The prototype pre-processes the Malay Qur’an and an input query using a stemming algorithm and then searches for occurrences of the query word stem. The result produced shows improved matching likelihood between user query and its answer. A POS-tagging algorithm has also been produced. The stemming and tagging algorithms can be used as tools for research related to other Malay texts and can be used to support applications such as information retrieval, question answering systems, ontology-based search and other text analysis tasks.

Keywords: language resource, Malay translated Qur'an, natural language processing (NLP), text processing

Procedia PDF Downloads 281

Authors: Victor N. Anisimov, Lyubov A. Boyko, Yazgul R. Almukhametova, Natalia V. Galkina, Alexander V. Latanov

Abstract:

Modern commercials presented on billboards, TV and on the Internet contain a lot of information about the product or service in text form. However, this information cannot always be perceived and understood by consumers. Typical sociological focus group studies often cannot reveal important features of the interpretation and understanding information that has been read in text messages. In addition, there is no reliable method to determine the degree of understanding of the information contained in a text. Only the fact of viewing a text does not mean that consumer has perceived and understood the meaning of this text. At the same time, the tools based on marketing analysis allow only to indirectly estimate the process of reading and understanding a text. Therefore, the aim of this work is to develop a valid method of recording objective indicators in real time for assessing the fact of reading and the degree of text comprehension. Psychophysiological parameters recorded during text reading can form the basis for this objective method. We studied the relationship between multimodal psychophysiological parameters and the process of text comprehension during reading using the method of correlation analysis. We used eye-tracking technology to record eye movements parameters to estimate visual attention, electroencephalography (EEG) to assess cognitive load and polygraphic indicators (skin-galvanic reaction, SGR) that reflect the emotional state of the respondent during text reading. We revealed reliable interrelations between perceiving the information and the dynamics of psychophysiological parameters during reading the text in commercials. Eye movement parameters reflected the difficulties arising in respondents during perceiving ambiguous parts of text. EEG dynamics in rate of alpha band were related with cumulative effect of cognitive load. SGR dynamics were related with emotional state of the respondent and with the meaning of text and type of commercial. EEG and polygraph parameters together also reflected the mental difficulties of respondents in understanding text and showed significant differences in cases of low and high text comprehension. We also revealed differences in psychophysiological parameters for different type of commercials (static vs. video, financial vs. cinema vs. pharmaceutics vs. mobile communication, etc.). Conclusions: Our methodology allows to perform multimodal evaluation of text perusal and the quality of text reading in commercials. In general, our results indicate the possibility of designing an integral model to estimate the comprehension of reading the commercial text in percent scale based on all noticed markers.

Keywords: reading, commercials, eye movements, EEG, polygraphic indicators

Procedia PDF Downloads 137

Authors: Zayar Phyo, Ei Chaw Htoon

Abstract:

In a New-LSB based Steganography method, the Fisher-Yates algorithm is used to permute an existing array randomly. However, that algorithm performance became slower and occurred memory overflow problem while processing the large dimension of images. Therefore, the Text-Based Shuffling algorithm aimed to select only necessary pixels as hiding characters at the specific position of an image according to the length of the input text. In this paper, the enhanced text-based shuffling algorithm is presented with the powered of GPU to improve more excellent performance. The proposed algorithm employs the OpenCL Aparapi framework, along with XORShift Kernel including the Pseudo-Random Number Generator (PRNG) Kernel. PRNG is applied to produce random numbers inside the kernel of OpenCL. The experiment of the proposed algorithm is carried out by practicing GPU that it can perform faster-processing speed and better efficiency without getting the disruption of unnecessary operating system tasks.

Keywords: LSB based steganography, Fisher-Yates algorithm, text-based shuffling algorithm, OpenCL, XORShiftKernel

Procedia PDF Downloads 119

Authors: Takayuki Konishi, Kakeru Yazawa, Mariko Kondo

Abstract:

This study examines developmental change in the production of epenthetic vowels by Japanese learners of English in relation to acquisition of L2 English speech rhythm. Seventy-two Japanese learners of English in the J-AESOP corpus were divided into lower- and higher-level learners according to their proficiency score and the frequency of vowel epenthesis. Three learners were excluded because no vowel epenthesis was observed in their utterances. The analysis of their read English speech data showed no statistical difference between lower- and higher-level learners, implying the absence of any developmental change in durations of epenthetic vowels. This result, together with the findings of previous studies, will be discussed in relation to the transfer of L1 phonology and manifestation of L2 English rhythm.

Keywords: vowel epenthesis, Japanese learners of English, L2 speech corpus, speech rhythm

Procedia PDF Downloads 240

Authors: Sidi Yang, Haiyi Zhang

Abstract:

Twitter is a microblogging platform, where millions of users daily share their attitudes, views, and opinions. Using a probabilistic Latent Dirichlet Allocation (LDA) topic model to discern the most popular topics in the Twitter data is an effective way to analyze a large set of tweets to find a set of topics in a computationally efficient manner. Sentiment analysis provides an effective method to show the emotions and sentiments found in each tweet and an efficient way to summarize the results in a manner that is clearly understood. The primary goal of this paper is to explore text mining, extract and analyze useful information from unstructured text using two approaches: LDA topic modelling and sentiment analysis by examining Twitter plain text data in English. These two methods allow people to dig data more effectively and efficiently. LDA topic model and sentiment analysis can also be applied to provide insight views in business and scientific fields.

Keywords: text mining, Twitter, topic model, sentiment analysis

Procedia PDF Downloads 149

Authors: Amaal Almigal

Abstract:

Technology has been argued to offer distinct advantages and benefits for teaching children with autism spectrum disorder (ASD) to communicate. One aspect of this technology is augmentative and alternative communication (AAC) systems such as picture exchange or speech generation devices. Whilst there has been significant progress in teaching these children to request their wants and needs with AAC, there remains a need for developing technologies that can really make a difference in teaching them to ask questions. iPad-based AAC can be effective for communication. However, the effectiveness of this type of AAC in teaching children to ask questions needs to be examined. Thus, in order to examine the effectiveness of iPad-based AAC in teaching children with ASD to ask questions, This research will test whether iPad leads to more learning than a traditional approach picture and text cards does. Two groups of children who use AAC will be taught to ask ‘What is it?’ questions. With the first group, low-tech AAC picture and text cards will be used, while an iPad-based AAC application called Proloquo2Go will be used with the second group. Interviews with teachers and parents will be conducted before and after the experiment. The children’s perspectives will also be considered. The initial outcomes of this research indicate that iPad can be an effective tool to help children with autism to ask questions.

Keywords: autism, communication, information, iPad, pictures, requesting

Procedia PDF Downloads 243

Authors: K. Rhea Devaiah, B. S. Premalatha

Abstract:

Background: Tonsillar Lingual sulcus is the area between the tonsils and the base of the tongue. The surgical resection of the lesions in the head and neck results in changes in speech and swallowing functions. The severity of the speech and swallowing problem depends upon the site and extent of the lesion, types and extent of surgery and also the flexibility of the remaining structures. Need of the study: This paper focuses on the importance of speech and swallowing rehabilitation in an individual with the lesion in the Tonsillar Lingual Sulcus and post-operative functions. Aim: Evaluating the speech and swallow functions post-intensive speech and swallowing rehabilitation. The objectives are to evaluate the speech intelligibility and swallowing functions after intensive therapy and assess the quality of life. Method: The present study describes a report of an individual aged 47years male, with the diagnosis of basaloid squamous cell carcinoma, left tonsillar lingual sulcus (pT2n2M0) and underwent wide local excision with left radical neck dissection with PMMC flap reconstruction. Post-surgery the patient came with a complaint of reduced speech intelligibility, and difficulty in opening the mouth and swallowing. Detailed evaluation of the speech and swallowing functions were carried out such as OPME, articulation test, speech intelligibility, different phases of swallowing and trismus evaluation. Self-reported questionnaires such as SHI-E(Speech handicap Index- Indian English), DHI (Dysphagia handicap Index) and SESEQ -K (Self Evaluation of Swallowing Efficiency in Kannada) were also administered to know what the patient feels about his problem. Based on the evaluation, the patient was diagnosed with pharyngeal phase dysphagia associated with trismus and reduced speech intelligibility. Intensive speech and swallowing therapy was advised weekly twice for the duration of 1 hour. Results: Totally the patient attended 10 intensive speech and swallowing therapy sessions. Results indicated misarticulation of speech sounds such as lingua-palatal sounds. Mouth opening was restricted to one finger width with difficulty chewing, masticating, and swallowing the bolus. Intervention strategies included Oro motor exercise, Indirect swallowing therapy, usage of a trismus device to facilitate mouth opening, and change in the food consistency to help to swallow. A practice session was held with articulation drills to improve the production of speech sounds and also improve speech intelligibility. Significant changes in articulatory production and speech intelligibility and swallowing abilities were observed. The self-rated quality of life measures such as DHI, SHI and SESE Q-K revealed no speech handicap and near-normal swallowing ability indicating the improved QOL after the intensive speech and swallowing therapy. Conclusion: Speech and swallowing therapy post carcinoma in the tonsillar lingual sulcus is crucial as the tongue plays an important role in both speech and swallowing. The role of Speech-language and swallowing therapists in oral cancer should be highlighted in treating these patients and improving the overall quality of life. With intensive speech-language and swallowing therapy post-surgery for oral cancer, there can be a significant change in the speech outcome and swallowing functions depending on the site and extent of lesions which will thereby improve the individual’s QOL.

Keywords: oral cancer, speech and swallowing therapy, speech intelligibility, trismus, quality of life

Procedia PDF Downloads 77

Authors: Beier Zhu, Rui Zhang, Qi Song

Abstract:

Text contained within fixed-layout documents can be of great semantic value and so requires a high localization accuracy, such as ID cards, invoices, cheques, and passports. Recently, algorithms based on deep convolutional networks achieve high performance on text detection tasks. However, for text localization in fixed-layout documents, such algorithms detect word bounding boxes individually, which ignores the layout information. This paper presents a novel architecture built on convolutional neural networks (CNNs). A global text localization network and a regional bounding-box regression network are introduced to tackle the problem in a coarse-to-fine manner. The text localization network simultaneously locates word bounding points, which takes the layout information into account. The bounding-box regression network inputs the features pooled from arbitrarily sized RoIs and refine the localizations. These two networks share their convolutional features and are trained jointly. A typical type of fixed-layout documents: ID cards, is selected to evaluate the effectiveness of the proposed system. These networks are trained on data cropped from nature scene images, and synthetic data produced by a synthetic text generation engine. Experiments show that our approach locates high accuracy word bounding boxes and achieves state-of-the-art performance.

Keywords: bounding box regression, convolutional networks, fixed-layout documents, text localization

Procedia PDF Downloads 164

Authors: Kseniia Fedorova

Abstract:

The article is devoted to interlinguistic homonymy and enantiosemy analysis. These phenomena belong to the process of linguistic interference, which leads to violation of the communicative utterances integrity and causes misunderstanding between foreign interlocutors - native speakers of different Slavic languages. More attention is paid to investigation of non-typical speech situations, which occurred spontaneously or created by somebody intentionally being based on described phenomenon mechanism. The classification of typical students' mistakes connected with the paradox of interference is being represented in the article. The survey contributes to speech act theory, contemporary linguodidactics, translation science and comparative lexicology of Slavonic languages.

Keywords: adherent enantiosemy, interference, interslavonic homonymy, speech act

Procedia PDF Downloads 213

Authors: Bubai Maji, Monorama Swain

Abstract:

Speech is considered an essential and most natural medium for the interaction between machines and humans. However, extracting effective features for speech emotion recognition (SER) is remains challenging. The present studies show that the temporal information captured but high-level temporal-feature learning is yet to be investigated. In this paper, we present an efficient novel method using the Self-attention (SA) mechanism in a combination of Convolutional Neural Network (CNN) and Bi-directional Gated Recurrent Unit (Bi-GRU) network to learn high-level temporal-feature. In order to further enhance the representation of the high-level temporal-feature, we integrate a Bi-GRU output with learnable weights features by SA, and improve the performance. We evaluate our proposed method on our created SITB-OSED and IEMOCAP databases. We report that the experimental results of our proposed method achieve state-of-the-art performance on both databases.

Keywords: Bi-GRU, 1D-CNNs, self-attention, speech emotion recognition

Procedia PDF Downloads 89

Authors: Rabi Mouhcine, Amrouch Mustapha, Mahani Zouhir, Mammass Driss

Abstract:

In this paper, we present a system for offline recognition cursive Arabic handwritten text based on Hidden Markov Models (HMMs). The system is analytical without explicit segmentation used embedded training to perform and enhance the character models. Extraction features preceded by baseline estimation are statistical and geometric to integrate both the peculiarities of the text and the pixel distribution characteristics in the word image. These features are modelled using hidden Markov models and trained by embedded training. The experiments on images of the benchmark IFN/ENIT database show that the proposed system improves recognition.

Keywords: recognition, handwriting, Arabic text, HMMs, embedded training

Procedia PDF Downloads 322

Authors: Mahmoud al-Ashiriy

Abstract:

This paper begins from the idea that the real history of literature is the history of its style. And since the rhyme –as known- is not merely the last letter, that have received a lot of analysis and investigation, but it is a collection of other values in addition to its different markings. This paper will explore the work of the connecting ha’ and its effectiveness in shaping the text of poetry, since it establishes vocal rhythms in addition to its role in indicating references through the pronoun, vertically through the poem through the sequence of its verses, also horizontally through what environs the one verse of sentences. If the scientific formation of prosody stopped at the possibilities and prohibitions; literary criticism and poetry studies should explore what is above the rule of aesthetic horizon of poetic effectiveness that varies from a text to another, a poet to another, a literary period to another, or from a poetic taste to another. Then the paper will explore this poetic essence in the texts of the famous Andalusian Poet Al-Husari Al-Qayrawani through his well-known Daliyya (a poem that its verses end with the letter D), and the role of the connecting ha’ in fulfilling its text and the accomplishment of its poetics, departing from this to the diwan (the big collection of poems) also as a higher text that surpasses the text/poem, and through what it represents of effectiveness the work of the phenomenon in accomplishing the poetics of the poem of Al-Husari Al-Qayrawani who is one of the pillars of Arabic poetics in Andalusia.

Keywords: Al-Husari Al-Qayrawni, poetics, rhyme, stylistics, science of the text

Procedia PDF Downloads 531

Authors: Ming Liu, Chong Wu, Bingquan Liu, Lei Chen

Abstract:

Internet users have to face the massive amount of textual data every day. Organizing texts into categories can help users dig the useful information from large-scale text collection. Clustering, in fact, is one of the most promising tools for categorizing texts due to its unsupervised characteristic. Unfortunately, most of traditional clustering algorithms lose their high qualities on large-scale text collection. This situation mainly attributes to the high- dimensional vectors generated from texts. To effectively and efficiently cluster large-scale text collection, this paper proposes a vector reconstruction based clustering algorithm. Only the features that can represent the cluster are preserved in cluster’s representative vector. This algorithm alternately repeats two sub-processes until it converges. One process is partial tuning sub-process, where feature’s weight is fine-tuned by iterative process. To accelerate clustering velocity, an intersection based similarity measurement and its corresponding neuron adjustment function are proposed and implemented in this sub-process. The other process is overall tuning sub-process, where the features are reallocated among different clusters. In this sub-process, the features useless to represent the cluster are removed from cluster’s representative vector. Experimental results on the three text collections (including two small-scale and one large-scale text collections) demonstrate that our algorithm obtains high quality on both small-scale and large-scale text collections.

Keywords: vector reconstruction, large-scale text clustering, partial tuning sub-process, overall tuning sub-process

Procedia PDF Downloads 403

Authors: Rim Messaoudi, Nogaye-Gueye Gning, François Azelart

Abstract:

Automatic text classification applies mostly natural language processing (NLP) and other AI-guided techniques to automatically classify text in a faster and more accurate manner. This paper discusses the subject of using predictive maintenance to manage incident tickets inside the sociality. It focuses on proposing a tool that treats and analyses comments and notes written by administrators after resolving an incident ticket. The goal here is to increase the quality of these comments. Additionally, this tool is based on NLP and machine learning techniques to realize the textual analytics of the extracted data. This approach was tested using real data taken from the French National Railways (SNCF) company and was given a high-quality result.

Keywords: machine learning, text classification, NLP techniques, semantic representation

Procedia PDF Downloads 59

Authors: Paolo Fantozzi, Luigi Laura, Umberto Nanni

Abstract:

The problem of Entity relation discovery in structured data, a well covered topic in literature, consists in searching within unstructured sources (typically, text) in order to find connections among entities. These can be a whole dictionary, or a specific collection of named items. In many cases machine learning and/or text mining techniques are used for this goal. These approaches might be unfeasible in computationally challenging problems, such as processing massive data streams. A faster approach consists in collecting the cooccurrences of any two words (entities) in order to create a graph of relations - a cooccurrence graph. Indeed each cooccurrence highlights some grade of semantic correlation between the words because it is more common to have related words close each other than having them in the opposite sides of the text. Some authors have used sliding windows for such problem: they count all the occurrences within a sliding windows running over the whole text. In this paper we generalise such technique, coming up to a Weighted-Distance Sliding Window, where each occurrence of two named items within the window is accounted with a weight depending on the distance between items: a closer distance implies a stronger evidence of a relationship. We develop an experiment in order to support this intuition, by applying this technique to a data set consisting in the text of the Bible, split into verses.

Keywords: cooccurrence graph, entity relation graph, unstructured text, weighted distance

Procedia PDF Downloads 114

Authors: Aishwarya Talapuru, Sri Silpa Padmanabhuni, B. Jyoshna

Abstract:

Cryptography helps in preventing threats to information security by providing various algorithms. This study introduces a new symmetric key encryption algorithm for information security which is linked with the "raagas" which means Indian traditional scale and pattern of music notes. This algorithm takes the plain text as input and starts its encryption process. The algorithm then randomly selects a raaga from the list of raagas that is assumed to be present with both sender and the receiver. The plain text is associated with the thus selected raaga and an intermediate cipher-text is formed as the algorithm converts the plain text characters into other characters, depending upon the rules of the algorithm. This intermediate code or cipher text is arranged in various patterns in three different rounds of encryption performed. The total number of rounds in the algorithm is equal to the multiples of 3. To be more specific, the outcome or output of the sequence of first three rounds is again passed as the input to this sequence of rounds recursively, till the total number of rounds of encryption is performed. The raaga selected by the algorithm and the number of rounds performed will be specified at an arbitrary location in the key, in addition to important information regarding the rounds of encryption, embedded in the key which is known by the sender and interpreted only by the receiver, thereby making the algorithm hack proof. The key can be constructed of any number of bits without any restriction to the size. A software application is also developed to demonstrate this process of encryption, which dynamically takes the plain text as input and readily generates the cipher text as output. Therefore, this algorithm stands as one of the strongest tools for information security.

Keywords: cipher text, cryptography, plaintext, raaga

Procedia PDF Downloads 257

Authors: Armita Ghobadi, Samantha Emerson, Seyda Ozcaliskan

Abstract:

Being a bilingual involves mastery of both speech and gesture patterns in a second language (L2). We know from earlier work in first language (L1) production contexts that speech and co-speech gesture form a tightly integrated system: co-speech gesture mirrors the patterns observed in speech, suggesting an online effect of language on nonverbal representation of events in gesture during the act of speaking (i.e., “thinking for speaking”). Relatively less is known about the online effect of language on gesture in bilinguals speaking structurally different languages. The few existing studies—mostly with small sample sizes—suggests inconclusive findings: some show greater achievement of L2 patterns in gesture with more advanced L2 speech production, while others show preferences for L1 gesture patterns even in advanced bilinguals. In this study, we focus on advanced bilingual speakers of two structurally different languages (Spanish L1 with English L2) in comparison to L1 English speakers. We ask whether bilingual speakers will follow target L2 patterns not only in speech but also in gesture, or alternatively, follow L2 patterns in speech but resort to L1 patterns in gesture. We examined this question by studying speech and gestures produced by 23 advanced adult Spanish (L1)-English (L2) bilinguals (Mage=22; SD=7) and 23 monolingual English speakers (Mage=20; SD=2). Participants were shown 16 animated motion event scenes that included distinct manner and path components (e.g., "run over the bridge"). We recorded and transcribed all participant responses for speech and segmented it into sentence units that included at least one motion verb and its associated arguments. We also coded all gestures that accompanied each sentence unit. We focused on motion event descriptions as it shows strong crosslinguistic differences in the packaging of motion elements in speech and co-speech gesture in first language production contexts. English speakers synthesize manner and path into a single clause or gesture (he runs over the bridge; running fingers forward), while Spanish speakers express each component separately (manner-only: el corre=he is running; circle arms next to body conveying running; path-only: el cruza el puente=he crosses the bridge; trace finger forward conveying trajectory). We tallied all responses by group and packaging type, separately for speech and co-speech gesture. Our preliminary results (n=4/group) showed that productions in English L1 and Spanish L1 differed, with greater preference for conflated packaging in L1 English and separated packaging in L1 Spanish—a pattern that was also largely evident in co-speech gesture. Bilinguals’ production in L2 English, however, followed the patterns of the target language in speech—with greater preference for conflated packaging—but not in gesture. Bilinguals used separated and conflated strategies in gesture in roughly similar rates in their L2 English, showing an effect of both L1 and L2 on co-speech gesture. Our results suggest that online production of L2 language has more limited effects on L2 gestures and that mastery of native-like patterns in L2 gesture might take longer than native-like L2 speech patterns.

Keywords: bilingualism, cross-linguistic variation, gesture, second language acquisition, thinking for speaking hypothesis

Procedia PDF Downloads 46

Authors: Hussain Hameed Mayuuf

Abstract:

The study is an attempt to investigate the conceptual metonymies is used in political discourse about COVID-19. Thus, this study tries to analyze and investigate how the conceptual metonymies in Johnson's speech about coronavirus are constructed. This study aims at: Identifying how are metonymies relevant to understand the messages in Boris Johnson speeches and to find out how can conceptual blending theory help people to understand the messages in the political speech about COVID-19. Lastly, it tries to Point out the kinds of integration networks are common in political speech. The study is based on the hypotheses that conceptual blending theory is a powerful tool for investigating the intended messages in Johnson's speech and there are different processes of blending networks and conceptual mapping that enable the listeners to identify the messages in political speech. This study presents a qualitative and quantitative analysis of four speeches about COVID-19; they are said by Boris Johnson. The selected data have been tackled from the cognitive-semantic perspective by adopting Conceptual Blending Theory as a model for the analysis. It concludes that CBT is applicable to the analysis of metonymies in political discourse. Its mechanisms enable listeners to analyze and understand these speeches. Also the listener can identify and understand the hidden messages in Biden and Johnson's discourse about COVID-19 by using different conceptual networks. Finally, it is concluded that the double scope networks are the most common types of blending of metonymies in the political speech.

Keywords: cognitive, semantics, conceptual, metonymical, Covid-19

Procedia PDF Downloads 77

Authors: Arvind Aggarwal

Abstract:

We have developed a comprehensive global healthcare delivery System based on information technology. It has medical consultation system where a virtual consultant can give medical consultation to the patients and Doctors at the digital medical centre after reviewing the patient’s EMR file consisting of patient’s history, investigations in the voice, images and data format. The system has the surgical operation system too, where a remote robotic consultant can conduct surgery at the robotic surgical centre. The instant speech and text translation is incorporated in the software where the patient’s speech and text (language) can be translated into the consultant’s language and vice versa. A consultant of any specialty (surgeon or Physician) based in any country can provide instant health care consultation, to any patient in any country without loss of time. Robotic surgeons based in any country in a tertiary care hospital can perform remote robotic surgery, through patient friendly telemedicine and tele-surgical centres. The patient EMR, financial data and data of all the consultants and robotic surgeons shall be stored in cloud. It is a complete comprehensive business model with healthcare medical and surgical delivery system. The whole system is self-financing and can be implemented in any country. The entire system uses paperless, filmless techniques. This eliminates the use of all consumables thereby reduces substantial cost which is incurred by consumables. The consultants receive virtual patients, in the form of EMR, thus the consultant saves time and expense to travel to the hospital to see the patients. The consultant gets electronic file ready for reporting & diagnosis. Hence time spent on the physical examination of the patient is saved, the consultant can, therefore, spend quality time in studying the EMR/virtual patient and give his instant advice. The time consumed per patient is reduced and therefore can see more number of patients, the cost of the consultation per patients is therefore reduced. The additional productivity of the consultants can be channelized to serve rural patients devoid of doctors.

Keywords: e-health, telemedicine, telecare, IT-based healthcare

Procedia PDF Downloads 148