Search results for: speech processing

Authors: S. B. Rathna Kumar, Pranjali A Ujwane, Panchanan Mohanty

Abstract:

The present study aimed to develop a battery for assessing speech recognition performance by adults in Marathi. A total of four word lists were developed by considering word frequency, word familiarity, words in common use, and phonemic balance. Each word list consists of 25 words (15 monosyllabic words in CVC structure and 10 monosyllabic words in CVCV structure). Equivalence analysis and performance-intensity function testing was carried using the four word lists on a total of 150 native speakers of Marathi belonging to different regions of Maharashtra (Vidarbha, Marathwada, Khandesh and Northern Maharashtra, Pune, and Konkan). The subjects were further equally divided into five groups based on above mentioned regions. It was found that there was no significant difference (p > 0.05) in the speech recognition performance between groups for each word list and between word lists for each group. Hence, the four word lists developed were equally difficult for all the groups and can be used interchangeably. The performance-intensity (PI) function curve showed semi-linear function, and the groups’ mean slope of the linear portions of the curve indicated an average linear slope of 4.64%, 4.73%, 4.68%, and 4.85% increase in word recognition score per dB for list 1, list 2, list 3 and list 4 respectively. Although, there is no data available on speech recognition tests for adults in Marathi, most of the findings of the study are in line with the findings of research reports on other languages. The four word lists, thus developed, were found to have sufficient reliability and validity in assessing speech recognition performance by adults in Marathi.

Keywords: speech recognition performance, phonemic balance, equivalence analysis, performance-intensity function testing, reliability, validity

Procedia PDF Downloads 354

Authors: Deepthy Ann Joy, N. Sreedevi

Abstract:

Hearing impairment (HI) at an early age, identified before the onset of language development can reduce the negative effect on speech and language development of children. Early rehabilitation is very important in the improvement of speech production in children with HI. Other than conventional hearing aids, Cochlear Implants are being used in the rehabilitation of children with HI. However, delay in acquisition of speech and language milestones persist in children with Cochlear Implant (CI). Delay in speech milestones are reflected through speech sound errors. These errors reflect the temporal and spectral characteristics of speech. Hence, acoustical analysis of the speech sounds will provide a better representation of speech production skills in children with CI. The present study aimed at investigating the acoustic characteristics of vowels in Malayalam speaking children with a cochlear implant. The participants of the study consisted of 20 Malayalam speaking children in the age range of four and seven years. The experimental group consisted of 10 children with CI, and the control group consisted of 10 typically developing children. Acoustic analysis was carried out for 5 short (/a/, /i/, /u/, /e/, /o/) and 5 long vowels (/a:/, /i:/, /u:/, /e:/, /o:/) in word-initial position. The responses were recorded and analyzed for acoustic parameters such as Vowel duration, Ratio of the duration of a short and long vowel, Formant frequencies (F₁ and F₂) and Formant Centralization Ratio (FCR) computed using the formula (F₂u+F₂a+F₁i+F₁u)/(F₂i+F₁a). Findings of the present study indicated that the values for vowel duration were higher in experimental group compared to the control group for all the vowels except for /u/. Ratio of duration of short and long vowel was also found to be higher in experimental group compared to control group except for /i/. Further F₁ for all vowels was found to be higher in experimental group with variability noticed in F₂ values. FCR was found be higher in experimental group, indicating vowel centralization. Further, the results of independent t-test revealed no significant difference across the parameters in both the groups. It was found that the spectral and temporal measures in children with CI moved towards normal range. The result emphasizes the significance of early rehabilitation in children with hearing impairment. The role of rehabilitation related aspects are also discussed in detail which can be clinically incorporated for the betterment of speech therapeutic services in children with CI.

Keywords: acoustics, cochlear implant, Malayalam, vowels

Procedia PDF Downloads 143

Authors: Rahul Mohite, Lahcen Ouarbya

Abstract:

This paper focuses on the interplay between humans and computer systems and the ability of these systems to understand and respond to human emotions, including non-verbal communication. Current emotion recognition systems are based solely on either facial or verbal expressions. The limitation of these systems is that it requires large training data sets. The paper proposes a system for recognizing human emotions that combines both speech and emotion recognition. The system utilizes advanced techniques such as deep learning and image recognition to identify facial expressions and comprehend emotions. The results show that the proposed system, based on the combination of facial expression and speech, outperforms existing ones, which are based solely either on facial or verbal expressions. The proposed system detects human emotion with an accuracy of 86%, whereas the existing systems have an accuracy of 70% using verbal expression only and 76% using facial expression only. In this paper, the increasing significance and demand for facial recognition technology in emotion recognition are also discussed.

Keywords: facial reputation, expression reputation, deep gaining knowledge of, photo reputation, facial technology, sign processing, photo type

Procedia PDF Downloads 117

Authors: B. Afsharipoor, E. Nazemi

Abstract:

Artificial neural networks (ANNs) are powerful tool for prediction which can be trained based on a set of examples and thus, it would be useful for nonlinear image processing. The present paper reviews several paper regarding applications of ANN in image processing to shed the light on advantage and disadvantage of ANNs in this field. Different steps in the image processing chain including pre-processing, enhancement, segmentation, object recognition, image understanding and optimization by using ANN are summarized. Furthermore, results on using multi artificial neural networks are presented.

Keywords: neural networks, image processing, segmentation, object recognition, image understanding, optimization, MANN

Procedia PDF Downloads 404

Authors: Kais A. Kadhim

Abstract:

This paper is a semantic analysis of the English modals in Obama’s speech. The main objective of this study is to analyze selected modal auxiliaries identified in selected speeches of Obama’s campaign based on Coates’ (1983) semantic clusters. A total of fifteen speeches of Obama’s campaign were selected as the primary data and the modal auxiliaries selected for analysis include will, would, can, could, should, must, ought, shall, may and might. All the modal auxiliaries taken from the speeches of Barack Obama were analyzed based on the framework of Coates’ semantic clusters. Such analytical framework was carried out to examine how modal auxiliaries are used in the context of persuading people in Obama’s campaign speeches. The findings reveal that modals of intention, prediction, futurity and modals of possibility, ability, permission are mostly used in Obama’s campaign speeches.

Keywords: modals, meaning, persuasion, speech

Procedia PDF Downloads 405

Authors: Ahmed Elgadri

Abstract:

In the last thirty years, studies on cross-cultural pragmatics in general and apology strategies in specific have focused on western and East-Asian societies. A small volume of research has been conducted in investigating speech acts production by Arabic dialect speakers. Therefore, this study investigated the apology strategies used by Libyan Arabic speakers using an online Discourse Completion Task (DCT) questionnaire. The DCT consisted of six situations covering different social contexts. The survey was written in Libyan Arabic dialect to help generate vernacular speech as much as possible. The participants were 25 Libyan nationals, 12 females, and 13 males. Also, to get a deeper understanding of the motivation behind the use of certain strategies, the researcher interviewed four participants using the Libyan Arabic dialect as well. The results revealed a high use of IFID, offer of repair, and explanation. Although this might support the universality claim of speech acts strategies, it was clear that cultural norms and religion determined the choice of apology strategies significantly. This led to the discovery of new culture-specific strategies, as outlined later in this paper. This study gives an insight into politeness strategies in Libyan society, and it is hoped to contribute to the field of cross-cultural pragmatics.

Keywords: apologies, cross-cultural pragmatics, language and culture, Libyan Arabic, politeness, pragmatics, socio-pragmatics, speech acts

Procedia PDF Downloads 148

Authors: Muhammad Fairul Azreen Mohd Zahid

Abstract:

Speech, text, and movement intensify aspects of creating choreography by connecting with emotional entanglements, tradition, literature, and other texts. This research focuses on the practice as research that will prioritise the choreography process as an inquiry approach. With the driven context, the study intervenes in critical conjunctions of choreographic theory, bringing together new reflections on the moving body, spaces of action, as well as intertextuality between text and movements in making choreography. Throughout the process, the researcher will introduce the level of deliberation from speech through movements and text to express emotion within a narrative context of an “illocutionary act.” This practice as research will produce a different meaning from the “utterance text” to “utterance movements” in the perspective of speech acts theory by J.L Austin based on fragmented text from “pidato adat” which has been used as opening speech in Randai. Looking at the theory of deconstruction by Jacque Derrida also will give a different meaning from the text. Nevertheless, the process of creating the choreography will also help to lay the basic normative structure implicit in “constative” (statement text/movement) and “performative” (command text/movement). Through this process, the researcher will also look at several methods of using text from two works by Joseph Gonzales, “Becoming King-The Pakyung Revisited” and Crystal Pite's “The Statement,” as references to produce different methods in making choreography. The perspective from the semiotic foundation will support how occurrences within dance discourses as texts through a semiotic lens. The method used in this research is qualitative, which includes an interview and simulation of the concept to get an outcome.

Keywords: intertextuality, choreography, speech act, performative, deconstruction

Procedia PDF Downloads 96

Authors: Hajer Rahali, Zied Hajaiej, Noureddine Ellouze

Abstract:

The goal of speech parameterization is to extract the relevant information about what is being spoken from the audio signal. In speech recognition systems Mel-Frequency Cepstral Coefficients (MFCC) and Relative Spectral Mel-Frequency Cepstral Coefficients (RASTA-MFCC) are the two main techniques used. It will be shown in this paper that it presents some modifications to the original MFCC method. In our work the effectiveness of proposed changes to MFCC called Modified Function Cepstral Coefficients (MODFCC) were tested and compared against the original MFCC and RASTA-MFCC features. The prosodic features such as jitter and shimmer are added to baseline spectral features. The above-mentioned techniques were tested with impulsive signals under various noisy conditions within AURORA databases.

Keywords: auditory filter, impulsive noise, MFCC, prosodic features, RASTA filter

Procedia PDF Downloads 423

Authors: Djendi Mohmaed

Abstract:

In this paper, we propose a dual adaptive algorithm, which is based on the combination between the forward blind source separation (FBSS) structure and the affine projection algorithm (APA). This proposed algorithm combines the advantages of the source separation properties of the FBSS structure and the fast convergence characteristics of the APA algorithm. The proposed algorithm needs two noisy observations to provide an enhanced speech signal. This process is done in a blind manner without the need for ant priori information about the source signals. The proposed dual forward blind source separation affine projection algorithm is denoted (DFAPA) and used for the first time in an airplane cockpit context to enhance the communication from- and to- the airplane. Intensive experiments were carried out in this sense to evaluate the performance of the proposed DFAPA algorithm.

Keywords: adaptive algorithm, speech enhancement, system mismatch, SNR

Procedia PDF Downloads 133

Authors: Negar Riazifar, Nigel G. Stocks

Abstract:

This paper proposes strategies in level crossing (LC) sampling and reconstruction that provide high fidelity signal reconstruction for speech signals; these strategies circumvent the problem of exponentially increasing number of samples as the bit-depth is increased and hence are highly efficient. Specifically, the results indicate that the distribution of the intervals between samples is one of the key factors in the quality of signal reconstruction; including samples with short intervals do not improve the accuracy of the signal reconstruction, whilst samples with large intervals lead to numerical instability. The proposed sampling method, termed reduced conventional level crossing (RCLC) sampling, exploits redundancy between samples to improve the efficiency of the sampling without compromising performance. A reconstruction technique is also proposed that enhances the numerical stability through linear interpolation of samples separated by large intervals. Interpolation is demonstrated to improve the accuracy of the signal reconstruction in addition to the numerical stability. We further demonstrate that the RCLC and interpolation methods can give useful levels of signal recovery even if the average sampling rate is less than the Nyquist rate.

Keywords: level crossing sampling, numerical stability, speech processing, trigonometric polynomial

Procedia PDF Downloads 141

Authors: Abu Bakar Siddik

Abstract:

The July Movement in Bangladesh marked a pivotal moment in the nation’s struggle for democratic freedom and the right to free speech. This movement, driven by citizens, intellectuals, and activists, opposed authoritarian governance and the violation of civil liberties. By encouraging support for democratic reforms, it significantly changed the political landscape and highlighted the importance of grassroots activism for human rights. This essay examines the sociopolitical dynamics of the July Movement and its roots in popular resistance to authoritarian rule. It explores the movement's beginnings, emphasizing how citizens, scholars, and activists united to challenge the regime that restricted freedom of speech. In order to show how the movement gathered support for democratic reforms and ultimately helped bring about the overthrow of the regime, the article examines significant demonstrations, speeches, and government acts. This book offers a thorough examination of how the July Movement changed Bangladesh's political landscape by acting as a revolution for free speech and a trigger for the overthrow of autocratic authority, using historical documents, media coverage, and firsthand recollections. This study provides insightful information about how grassroots activism advances human rights.

Keywords: July movement, Bangladesh, free speech, democracy, authoritarianism, civil liberties, political change, human rights, social movements, protests, political landscape, regime change, activism, socio-political dynamics

Procedia PDF Downloads 14

Authors: Thomas Bryan, Veton Kepuska, Ivica Kostanic

Abstract:

A simple adaptive voice activity detector (VAD) is implemented using Gabor and gammatone atomic decomposition of speech for high Gaussian noise environments. Matching pursuit is used for atomic decomposition, and is shown to achieve optimal speech detection capability at high data compression rates for low signal to noise ratios. The most active dictionary elements found by matching pursuit are used for the signal reconstruction so that the algorithm adapts to the individual speakers dominant time-frequency characteristics. Speech has a high peak to average ratio enabling matching pursuit greedy heuristic of highest inner products to isolate high energy speech components in high noise environments. Gabor and gammatone atoms are both investigated with identical logarithmically spaced center frequencies, and similar bandwidths. The algorithm performs equally well for both Gabor and gammatone atoms with no significant statistical differences. The algorithm achieves 70% accuracy at a 0 dB SNR, 90% accuracy at a 5 dB SNR and 98% accuracy at a 20dB SNR using 30dB SNR as a reference for voice activity.

Keywords: atomic decomposition, gabor, gammatone, matching pursuit, voice activity detection

Procedia PDF Downloads 289

Authors: Mohammed Abu Shquier

Abstract:

This paper introduces a word segmentation method using the novel BP-LSTM-CRF architecture for processing semantic output training. The objective of web morphological analysis tools is to link a formal morpho-syntactic description to a lemma, along with morpho-syntactic information, a vocalized form, a vocalized analysis with morpho-syntactic information, and a list of paradigms. A key objective is to continuously enhance the proposed system through an inductive learning approach that considers semantic influences. The system is currently under construction and development based on data-driven learning. To evaluate the tool, an experiment on homograph analysis was conducted. The tool also encompasses the assumption of deep binary segmentation hypotheses, the arbitrary choice of trigram or n-gram continuation probabilities, language limitations, and morphology for both Modern Standard Arabic (MSA) and Dialectal Arabic (DA), which provide justification for updating this system. Most Arabic word analysis systems are based on the phonotactic morpho-syntactic analysis of a word transmitted using lexical rules, which are mainly used in MENA language technology tools, without taking into account contextual or semantic morphological implications. Therefore, it is necessary to have an automatic analysis tool taking into account the word sense and not only the morpho-syntactic category. Moreover, they are also based on statistical/stochastic models. These stochastic models, such as HMMs, have shown their effectiveness in different NLP applications: part-of-speech tagging, machine translation, speech recognition, etc. As an extension, we focus on language modeling using Recurrent Neural Network (RNN); given that morphological analysis coverage was very low in dialectal Arabic, it is significantly important to investigate deeply how the dialect data influence the accuracy of these approaches by developing dialectal morphological processing tools to show that dialectal variability can support to improve analysis.

Keywords: NLP, DL, ML, analyser, MSA, RNN, CNN

Procedia PDF Downloads 41

Authors: Shamm Kidane, Ibrahim Abdella, Fitsum Gaim, Simon Mulugeta, Sirak Asmerom, Natnael Ambasager, Yoel Ghebrihiwot

Abstract:

The call for natural language processing (NLP) systems for low-resource languages has become more apparent than ever in the past few years, with the arduous challenges still present in preparing such systems. This paper presents an improved dataset version of the Nagaoka Tigrinya Corpus for Parts-of-Speech (POS) classification system in the Tigrinya language. The size of the initial Nagaoka dataset was incremented, totaling the new tagged corpus to 118K tokens, which comprised the 12 basic POS annotations used previously. The additional content was also annotated manually in a stringent manner, followed similar rules to the former dataset and was formatted in CONLL format. The system made use of the novel approach in NLP tasks and use of the monolingually pre-trained TiELECTRA, TiBERT and TiRoBERTa transformer models. The highest achieved score is an impressive weighted F1-score of 94.2%, which surpassed the previous systems by a significant measure. The system will prove useful in the progress of NLP-related tasks for Tigrinya and similarly related low-resource languages with room for cross-referencing higher-resource languages.

Keywords: Tigrinya POS corpus, TiBERT, TiRoBERTa, conditional random fields

Procedia PDF Downloads 99

Authors: Nataliya L. Ogurechnikova

Abstract:

The paper aims to clarify the function of articles in the English speech and specify their place and role in the English language, taking into account the use of articles for delimitation of speech. A focus of the paper is the use of the definite and the indefinite articles with different types of noun phrases which comprise either one noun with or without attributes, such as the King, the Queen, the Lion, the Unicorn, a dimple, a smile, a new language, an unknown dialect, or several nouns with or without attributes, such as the King and Queen of Hearts, the Lion and Unicorn, a dimple or smile, a completely isolated language or dialect. It is stated that the function of delimitation is related to perception: the number of speech units in a text correlates with the way the speaker perceives and segments the denotation. The two following combinations of words the house and garden and the house and the garden contain different numbers of speech units, one and two respectively, and reveal two different perception modes which correspond to the use of the definite article in the examples given. Thus, the function of delimitation is twofold, it is related to perception and cognition, on the one hand, and, on the other hand, to grammar, if the subject of grammar is the structure of speech. Analysis of speech units in the paper is not limited by noun phrases and is amplified by discussion of peripheral phenomena which are nevertheless important because they enable to qualify articles as a syntactic phenomenon whereas they are not infrequently described in terms of noun morphology. With this regard attention is given to the history of linguistic studies, specifically to the description of English articles by Niels Haislund, a disciple of Otto Jespersen. A discrepancy is noted between the initial plan of Jespersen who intended to describe articles as a syntactic phenomenon in ‘A Modern English Grammar on Historical Principles’ and the interpretation of articles in terms of noun morphology, finally given by Haislund. Another issue of the paper is correlation between description and denotation, being a traditional aspect of linguistic studies focused on articles. An overview of relevant studies, given in the paper, goes back to the works of G. Frege, which gave rise to a series of scientific works where the meaning of articles was described within the scope of logical semantics. Correlation between denotation and description is treated in the paper as the meaning of article, i.e. a component in its semantic structure, which differs from the function of delimitation and is similar to the meaning of other quantifiers. The paper further explains why the relation between description and denotation, i.e. the meaning of English article, is irrelevant for noun morphology and has nothing to do with nominal categories of the English language.

Keywords: delimitation of speech, denotation, description, perception, speech units, syntax

Procedia PDF Downloads 240

Authors: Sankalp Devanand, Prateek Agasimani, Shamith V. S., Rohith Neeraje

Abstract:

Machine translation has witnessed significant advancements in recent years, but the translation of languages with distinct linguistic characteristics, such as English and Sanskrit, remains a challenging task. This research presents the development of a dedicated English-to-Sanskrit machine translation model, aiming to bridge the linguistic and cultural gap between these two languages. Using a variety of natural language processing (NLP) approaches, including FastText embeddings, this research proposes a thorough method to improve word meaning retrieval. Data preparation, part-of-speech tagging, dictionary searches, and transliteration are all included in the methodology. The study also addresses the implementation of an interpreter pattern and uses a word similarity task to assess the quality of word embeddings. The experimental outcomes show how the suggested approach may be used to enhance word meaning retrieval tasks with greater efficacy, accuracy, and adaptability. Evaluation of the model's performance is conducted through rigorous testing, comparing its output against existing machine translation systems. The assessment includes quantitative metrics such as BLEU scores, METEOR scores, Jaccard Similarity, etc.

Keywords: machine translation, English to Sanskrit, natural language processing, word meaning retrieval, fastText embeddings

Procedia PDF Downloads 42

Authors: Anindita Banik, Anjali Kant, Aninda Duti Banik, Arun Banik

Abstract:

Speech is a complex, dynamic unique motor activity through which we express thoughts and emotions and respond to and control our environment. The aim was based to compare select Motor Speech parameters and their sub parameters across typical Marathi speaking adults and children. The subjects included a total of 300 divided into Group I, II, III including males and females. Subjects included were reported of no significant medical history and had a rating of 0-1 on GRBAS scale. The recordings were obtained utilizing three stimuli for the acoustic analysis of Diadochokinetic rate (DDK), Second Formant Transition, Voice and Tremor and its sub parameters. And these aforementioned parameters were acoustically analyzed in Motor Speech Profile software in VisiPitch IV. The statistical analyses were done by applying descriptive statistics and Two- Way ANOVA.The results obtained showed statistically significant difference across age groups and gender for the aforementioned parameters and its sub parameters.In DDK, for avp (ms) there was a significant difference only across age groups. However, for avr (/s) there was a significant difference across age groups and gender. It was observed that there was an increase in rate with an increase in age groups. The second formant transition sub parameter F2 magn (Hz) also showed a statistically significant difference across both age groups and gender. There was an increase in mean value with an increase in age. Females had a higher mean when compared to males. For F2 rate (/s) a statistically significant difference was observed across age groups. There was an increase in mean value with increase in age. It was observed for Voice and Tremor MFTR (%) that a statistically significant difference was present across age groups and gender. Also for RATR (Hz) there was statistically significant difference across both age groups and gender. In other words, the values of MFTR and RATR increased with an increase in age. Thus, this study highlights the variation of the motor speech parameters amongst the typical population which would be beneficial for comparison with the individuals with motor speech disorders for assessment and management.

Keywords: adult, children, diadochokinetic rate, second formant transition, tremor, voice

Procedia PDF Downloads 307

Authors: Khenfer-Koummich Fatima, Mesbahi Larbi, Hendel Fatiha

Abstract:

This paper presents a recognition system for isolated words like robot commands. It’s carried out by Time Delay Neural Networks; TDNN. To teleoperate a robot for specific tasks as turn, close, etc… In industrial environment and taking into account the noise coming from the machine. The choice of TDNN is based on its generalization in terms of accuracy, in more it acts as a filter that allows the passage of certain desirable frequency characteristics of speech; the goal is to determine the parameters of this filter for making an adaptable system to the variability of speech signal and to noise especially, for this the back propagation technique was used in learning phase. The approach was applied on commands pronounced in two languages separately: The French and Arabic. The results for two test bases of 300 spoken words for each one are 87%, 97.6% in neutral environment and 77.67%, 92.67% when the white Gaussian noisy was added with a SNR of 35 dB.

Keywords: TDNN, neural networks, noise, speech recognition

Procedia PDF Downloads 288

Authors: Olga A. Gnatyuk

Abstract:

The article is devoted to the phenomenon of interference that occurs in the case of the Russian-Spanish language contact. The questions of the definition of the term and levels, as well as prerequisites of interference occurrence, are considered. Interference, which is an essential part of bilingualism, may become apparent at different linguistic levels. Interference is especially evident in oral speech. The article reviews some examples of grammatical interference in Russian-Spanish bilingualism of Russian immigrants living in Spain. According to the results of the research, some cases of mother-tongue interference in Russian-Speaking Spanish language learners’ speech were revealed. Special attention is paid to such key spheres of grammatical interference as articles, personal pronouns, gender, and number of nouns. In the research, the drop of a link-verb, as well as its usage in some incorrect form, are observed in Russian immigrants’ speech. Conclusions are drawn that in the Spanish language, interference errors appear because of a consequence of both the absence in the Russian language of certain phenomena and categories of the Spanish language and the discrepancy of the linguistic systems of the two languages.

Keywords: bilingualism, interference, grammatical interference, Russian language, Spanish language

Procedia PDF Downloads 158

Authors: Tian Lan, Yixiang Wang, Wenxin Tai, Yilan Lyu, Zufeng Wu

Abstract:

The monaural speech enhancement task in the time-frequencydomain has a myriad of approaches, with the stacked con-volutional neural network (CNN) demonstrating superiorability in feature extraction and selection. However, usingstacked single convolutions method limits feature represen-tation capability and generalization ability. In order to solvethe aforementioned problem, we propose an attention-basedadaptive convolutional network that integrates the multi-scale convolutional operations into a operation-specific blockvia input dependent attention to adapt to complex auditoryscenes. In addition, we introduce a two-stage progressivelearning method to enlarge the receptive field without a dra-matic increase in computation burden. We conduct a series ofexperiments based on the TIMIT corpus, and the experimen-tal results prove that our proposed model is better than thestate-of-art models on all metrics.

Keywords: speech enhancement, adaptive convolu-tion, progressive learning, time-frequency domain

Procedia PDF Downloads 120

Authors: Bipasha Sen, Aditya Agarwal

Abstract:

Multilingual automatic speech recognition (ASR) system is a single entity capable of transcribing multiple languages sharing a common phone space. Performance of such a system is highly dependent on the compatibility of the languages. State of the art speech recognition systems are built using sequential architectures based on recurrent neural networks (RNN) limiting the computational parallelization in training. This poses a significant challenge in terms of time taken to bootstrap and validate the compatibility of multiple languages for building a robust multilingual system. Complex architectural choices based on self-attention networks are made to improve the parallelization thereby reducing the training time. In this work, we propose Reed, a simple system based on 1D convolutions which uses very short context to improve the training time. To improve the performance of our system, we use raw time-domain speech signals directly as input. This enables the convolutional layers to learn feature representations rather than relying on handcrafted features such as MFCC. We report improvement on training and inference times by atleast a factor of 4x and 7.4x respectively with comparable WERs against standard RNN based baseline systems on SpeechOcean's multilingual low resource dataset.

Keywords: convolutional neural networks, language compatibility, low resource languages, multilingual automatic speech recognition

Procedia PDF Downloads 122

Authors: K. R. Roopesh Bharatwaj, Avinash Maharana, Favour Tobi Aborisade, Roger Young

Abstract:

Presently, there are few cases of complete automation of drones and its allied intelligence capabilities. In essence, the potential of the drone has not yet been fully utilized. This paper presents feasible methods to build an intelligent drone with smart capabilities such as self-driving, and obstacle avoidance. It does this through advanced Reinforcement Learning Techniques and performs object detection using latest advanced algorithms, which are capable of processing light weight models with fast training in real time instances. For the scope of this paper, after researching on the various algorithms and comparing them, we finally implemented the Deep-Q-Networks (DQN) algorithm in the AirSim Simulator. In future works, we plan to implement further advanced self-driving and object detection algorithms, we also plan to implement voice-based speech recognition for the entire drone operation which would provide an option of speech communication between users (People) and the drone in the time of unavoidable circumstances. Thus, making drones an interactive intelligent Robotic Voice Enabled Service Assistant. This proposed drone has a wide scope of usability and is applicable in scenarios such as Disaster management, Air Transport of essentials, Agriculture, Manufacturing, Monitoring people movements in public area, and Defense. Also discussed, is the entire drone communication based on the satellite broadband Internet technology for faster computation and seamless communication service for uninterrupted network during disasters and remote location operations. This paper will explain the feasible algorithms required to go about achieving this goal and is more of a reference paper for future researchers going down this path.

Keywords: convolution neural network, natural language processing, obstacle avoidance, satellite broadband technology, self-driving

Procedia PDF Downloads 249

Authors: Kwan Hee Han, Jun Woo Lee, Sung Moon Bae, Twae Kyung Park

Abstract:

ECA (Event-Condition-Action) languages are largely adopted for event processing since they are an intuitive and powerful paradigm for programming reactive systems. However, there are some limitations about ECA rules for processing of complex events such as coupling of event producer and consumer. The objective of this paper is to propose an ECA rule pattern to improve the current limitations of ECA rule, and to develop a prototype system. In this paper, conventional ECA rule is separated into 3 parts and each part is extended to meet the requirements of CEP. Finally, event processing logic is established by combining the relevant elements of 3 parts. The usability of proposed extended ECA rule is validated by a test scenario in this study.

Keywords: complex event processing, ECA rule, Event processing system, event-driven architecture, internet of things

Procedia PDF Downloads 528

Authors: Firas Alfwaress, Abdelraheem Bebers Abdelhadi Hamasha, Maha Abu Awaad

Abstract:

Objectives: The aim of the present study was to investigate the effect of Palatal Lift Prosthesis (PLP) on speech clarity in patients with Flaccid Dysarthria. Five speech measures were investigated including Nasalance Scores, Diadchokinetic (DDK), Vowel Duration, airflow, and Sound Intensity. Participants: Twelve (7 Males and 5 females) native speakers of Jordanian Arabic with Flaccid Dysarthria following stroke, traumatic brain injury, and amyotrophic lateral sclerosis were included. The age of the participants ranged from 8–65 years with an average of 31.75 years. Design: Nasalance Scores, Diadchokinetic rate, Vowel Duration, and Sound Intensity were obtained using the Nasometer II, Model 6450 in three conditions. The first condition included obtaining the five measures without wearing the customized Palatal Lift Prosthesis. The second and third conditions included obtaining the five measures immediately after wearing the Palatal Lift Prosthesis and three months later. Results: Palatal lift prosthesis was found to be effective in individuals with flaccid dysarthria. Results showed decrease in the Nasalance Scores for the syllable repetition tasks and vowel prolongation tasks when comparing the means in the pre PLP with the post PLP at p≤0.001 except for the /m/ prolongation task. Results showed increased DDK repetition task, airflow amount, and sound intensity, and a decrease in vowel length at p≤0.001. Conclusions: The use of palatal lift prosthesis is effective in improving the speech of patients with flaccid dysarthria.

Keywords: palatal lift prosthesis, flaccid dysarthria, hypernasality, speech clarity, diadchokinetic rate

Procedia PDF Downloads 384

Authors: Boikanyego Sebina

Abstract:

The present study investigates the effects of socioeconomic status (SES) and bilingualism on the Setswana speech rhythm of Batswana (citizens) children aged 6-7 years with typical development born and residing in Botswana. Botswana is a country in which there is a diglossic Setswana/English language setting, where English is the dominant high-status language in educational and public contexts. Generally, children from low SES have lower linguistic and cognitive profiles than their age-matched peers from high SES. A greater understanding of these variables would allow educators to distinguish between underdeveloped language skills in children due to impairment and environmental issues for them to successfully enroll children in language development enhancement programs specific to the child’s needs. There are 20 participants: 10 high SES private English-medium educated early sequential Setswana-English bilingual children, taught full-time in English (L2) from the age of 3 years, and for whom English has become dominant; and 10 low SES children who are educated in public schools for whom English is considered a learner language, i.e., L1 Setswana is dominant. The aim is to see whether SES and bilingualism, have had an effect on the Setswana speech rhythm of children in either group. The study primarily uses semi-spontaneous speech based on the telling of the wordless picture storybook. A questionnaire is used to elicit the language use pattern of the children and that of their parents, as well as the education level of the parents and the school the children attend. A comparison of the rhythm shows that children from high SES have a lower durational variability than those from low SES. The findings of the study are that the low durational variability by children from high SES may suggest an underdeveloped rhythm. In conclusion, the results of the present study are against the notion that children from high SES outperform those from low SES in linguistic development.

Keywords: bilingualism, Setswana English, socio-economic status, speech-rhythm

Procedia PDF Downloads 66

Authors: Saeed Qaisrani

Abstract:

This article briefly reviews the rise of Critical Discourse Analysis about the Pakistani President Mamnoon Hussain speech which delivered in the joint session of Parliament and teases out a detailed analysis of the various critiques that have been levelled at CDA and its practitioners over the last twenty years, both by scholars working within the “critical” paradigm and by other critics. A range of criticisms are discussed which target the underlying premises, the analytical methodology and the disputed areas of reader response and the integration of contextual factors. Controversial issues such as the predominantly negative focus of much CDA scholarship, and the status of CDA as an emergent “intellectual orthodoxy”, are also reviewed. The conclusions offer a summary of the principal criticisms that emerge from this overview, and suggest some ways in which these problems could be attenuated. It also focused on the different views about president speech and how it is presented in the Pakistani print and electronic media.

Keywords: Critical Discourse Analysis, Analytical methodology, Corpus linguistics, Reader response theory, Critical paradigm, Contextualization.

Procedia PDF Downloads 483

Authors: Thein Thein, Kalyar Myo San

Abstract:

Lip reading system is one of the different supportive technologies for hearing impaired, or elderly people or non-native speakers. For normal hearing persons in noisy environments or in conditions where the audio signal is not available, lip reading techniques can be used to increase their understanding of spoken language. Hearing impaired persons have used lip reading techniques as important tools to find out what was said by other people without hearing voice. Thus, visual speech information is important and become active research area. Using visual information from lip movements can improve the accuracy and robustness of a speech recognition system and the need for lip reading system is ever increasing for every language. However, the recognition of lip movement is a difficult task because of the region of interest (ROI) is nonlinear and noisy. Therefore, this paper proposes method to detect the accurate lips shape and to localize lip movement towards automatic lip tracking by using the combination of Otsu global thresholding technique and Moore Neighborhood Tracing Algorithm. Proposed method shows how accurate lip localization and tracking which is useful for speech recognition. In this work of study and experiments will be carried out the automatic lip localizing the lip shape for Myanmar consonants using the only visual information from lip movements which is useful for visual speech of Myanmar languages.

Keywords: lip reading, lip localization, lip tracking, Moore neighborhood tracing algorithm

Procedia PDF Downloads 352

Authors: Vishal Raj, Noorhan Abbas

Abstract:

Ambiguity in NLP (Natural language processing) refers to the ability of a word, phrase, sentence, or text to have multiple meanings. This results in various kinds of ambiguities such as lexical, syntactic, semantic, anaphoric and referential am-biguities. This study is focused mainly on solving the issue of Lexical ambiguity. Word Sense Disambiguation (WSD) is an NLP technique that aims to resolve lexical ambiguity by determining the correct meaning of a word within a given context. Most WSD solutions rely on words for training and testing, but we have used lemma and Part of Speech (POS) tokens of words for training and testing. Lemma adds generality and POS adds properties of word into token. We have designed a novel method to create an affinity matrix to calculate the affinity be-tween any pair of lemma_POS (a token where lemma and POS of word are joined by underscore) of given training set. Additionally, we have devised an al-gorithm to create the sense clusters of tokens using affinity matrix under hierar-chy of POS of lemma. Furthermore, three different mechanisms to predict the sense of target word using the affinity/similarity value are devised. Each contex-tual token contributes to the sense of target word with some value and whichever sense gets higher value becomes the sense of target word. So, contextual tokens play a key role in creating sense clusters and predicting the sense of target word, hence, the model is named Contextual SenSe Model (CSM). CSM exhibits a noteworthy simplicity and explication lucidity in contrast to contemporary deep learning models characterized by intricacy, time-intensive processes, and chal-lenging explication. CSM is trained on SemCor training data and evaluated on SemEval test dataset. The results indicate that despite the naivety of the method, it achieves promising results when compared to the Most Frequent Sense (MFS) model.

Keywords: word sense disambiguation (wsd), contextual sense model (csm), most frequent sense (mfs), part of speech (pos), natural language processing (nlp), oov (out of vocabulary), lemma_pos (a token where lemma and pos of word are joined by underscore), information retrieval (ir), machine translation (mt)

Procedia PDF Downloads 105

Authors: Marija Zulić, Vanda Hájková, Nina Brkić–Jovanović, Srećko Potić, Sanja Tomić

Abstract:

The name cerebral palsy comes from the word cerebrum, which means the brain and the word palsy, which means seizure, and essentially refers to the movement disorder. In the clinical picture of cerebral palsy, basic neuromotor disorders are associated with other various disorders: behavioural, intellectual, speech, sensory, epileptic seizures, and bone and joint deformities. Motor speech disorders are among the most common difficulties present in people with cerebral palsy. Social participation represents an interaction between an individual and their social environment. Quality of social participation of the students with cerebral palsy at school is an important indicator of their successful participation in adulthood. One of the most important skills for the undisturbed social participation is ability of good communication. The aim of the study was to determine relation between social participation of students with cerebral palsy and presence of their speech impairment in primary schools in the Czech Republic. The study was performed in the Czech Republic in mainstream schools and schools established for the pupils with special education needs. We analysed 75 children with cerebral palsy aged between six and twelve years attending up to sixth grade by using the first and the third part of the school function assessment questionnaire as the main instrument. The other instrument we used in the research is the Gross motor function classification system–five–level classification system, which measures degree of motor functions of children and youth with cerebral palsy. Funding for this study was provided by the Grant Agency of Charles University in Prague.

Keywords: cerebral palsy, social participation, speech disorders, The Czech Republic, the school function assessment

Procedia PDF Downloads 283

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

Abstract:

Speech emotion classification is a dominant research field in finding a sturdy and profligate classifier appropriate for different real-life applications. This effort accentuates on classifying different emotions from speech signal quarried from the features related to pitch, formants, energy contours, jitter, shimmer, spectral, perceptual and temporal features. Tensor deep stacking neural networks were supported to examine the factors that influence the classification success rate. Facial electromyography signals were composed of several forms of focuses in a controlled atmosphere by means of audio-visual stimuli. Proficient facial electromyography signals were pre-processed using moving average filter, and a set of arithmetical features were excavated. Extracted features were mapped into consistent emotions using bilinear mapping. With facial electromyography signals, a database comprising diverse emotions will be exposed with a suitable fine-tuning of features and training data. A success rate of 92% can be attained deprived of increasing the system connivance and the computation time for sorting diverse emotional states.

Keywords: speech emotion classification, tensor deep stacking neural networks, facial electromyography, bilinear mapping, audio-visual stimuli

Procedia PDF Downloads 253