Search results for: L2 speech corpus

Authors: Pogula Rakesh, T. Kishore Kumar

Abstract:

Speech enhancement is a long standing problem with numerous applications like teleconferencing, VoIP, hearing aids, and speech recognition. The motivation behind this research work is to obtain a clean speech signal of higher quality by applying the optimal noise cancellation technique. Real-time adaptive filtering algorithms seem to be the best candidate among all categories of the speech enhancement methods. In this paper, we propose a speech enhancement method based on Recursive Least Squares (RLS) adaptive filter of speech signals. Experiments were performed on noisy data which was prepared by adding AWGN, Babble and Pink noise to clean speech samples at -5dB, 0dB, 5dB, and 10dB SNR levels. We then compare the noise cancellation performance of proposed RLS algorithm with existing NLMS algorithm in terms of Mean Squared Error (MSE), Signal to Noise ratio (SNR), and SNR loss. Based on the performance evaluation, the proposed RLS algorithm was found to be a better optimal noise cancellation technique for speech signals.

Keywords: adaptive filter, adaptive noise canceller, mean squared error, noise reduction, NLMS, RLS, SNR, SNR loss

Procedia PDF Downloads 483

Authors: Adel Assiri, Ahmed Emam, Hmood Al-Dossari

Abstract:

Sentiment analysis (SA) has received growing attention in Arabic language research. However, few studies have yet to directly apply SA to Arabic due to lack of a publicly available dataset for this language. This paper partially bridges this gap due to its focus on one of the Arabic dialects which is the Saudi dialect. This paper presents annotated data set of 4700 for Saudi dialect sentiment analysis with (K= 0.807). Our next work is to extend this corpus and creation a large-scale lexicon for Saudi dialect from the corpus.

Keywords: Arabic, sentiment analysis, Twitter, annotation

Procedia PDF Downloads 633

Authors: Jarek Krajewski, Andre Wittenborn, Martin Sauerland

Abstract:

This abstract describes a promising approach for estimating post-traumatic stress disorder (PTSD) based on prosodic speech characteristics. It illustrates the validity of this method by briefly discussing results from an Arabic refugee sample (N= 47, 32 m, 15 f). A well-established standardized self-report scale “Reaction of Adolescents to Traumatic Stress” (RATS) was used to determine the ground truth level of PTSD. The speech material was prompted by telling about autobiographical related sadness inducing experiences (sampling rate 16 kHz, 8 bit resolution). In order to investigate PTSD-induced speech changes, a self-developed set of 136 prosodic speech features was extracted from the .wav files. This set was adapted to capture traumatization related speech phenomena. An artificial neural network (ANN) machine learning model was applied to determine the PTSD level and reached a correlation of r = .37. These results indicate that our classifiers can achieve similar results to those seen in speech-based stress research.

Keywords: speech prosody, PTSD, machine learning, feature extraction

Procedia PDF Downloads 91

Authors: A. Belmeguenai, K. Mansouri, R. Djemili

Abstract:

This work present a new algorithm based on the nonlinear filter generator for speech encryption and decryption. The proposed algorithm consists on the use a linear feedback shift register (LFSR) whose polynomial is primitive and nonlinear Boolean function. The purpose of this system is to construct Keystream with good statistical properties, but also easily computable on a machine with limited capacity calculated. This proposed speech encryption scheme is very simple, highly efficient, and fast to implement the speech encryption and decryption. We conclude the paper by showing that this system can resist certain known attacks.

Keywords: nonlinear filter generator, stream ciphers, speech encryption, security analysis

Procedia PDF Downloads 297

Authors: XuKe Cao

Abstract:

This paper reviews the current state of research on low-resource languages in the field of speech recognition, focusing on the challenges faced by low-resource language speech recognition, including the scarcity of data resources, the lack of linguistic resources, and the diversity of dialects and accents. The article reviews recent progress in low-resource language speech recognition, including techniques such as data augmentation, end to-end models, transfer learning, and multi-task learning. Based on the challenges currently faced, the paper also provides an outlook on future research directions. Through these studies, it is expected that the performance of speech recognition for low resource languages can be improved, promoting the widespread application and adoption of related technologies.

Keywords: low-resource languages, speech recognition, data augmentation techniques, NLP

Procedia PDF Downloads 17

Authors: S. Jagadeesh Kumar

Abstract:

This expose presents a luculent of recent machine learning practices as employed in the modern and as pertinent to prospective automatic speech recognition schemes. The aspiration is to promote additional traverse ablution among the machine learning and automatic speech recognition factions that have transpired in the precedent. The manuscript is structured according to the chief machine learning archetypes that are furthermore trendy by now or have latency for building momentous hand-outs to automatic speech recognition expertise. The standards offered and convoluted in this article embraces adaptive and multi-task learning, active learning, Bayesian learning, discriminative learning, generative learning, supervised and unsupervised learning. These learning archetypes are aggravated and conferred in the perspective of automatic speech recognition tools and functions. This manuscript bequeaths and surveys topical advances of deep learning and learning with sparse depictions; further limelight is on their incessant significance in the evolution of automatic speech recognition.

Keywords: automatic speech recognition, deep learning methods, machine learning archetypes, Bayesian learning, supervised and unsupervised learning

Procedia PDF Downloads 448

Authors: Manjare Chandraprabha A., S. D. Shirbahadurkar, Manjare Anil S., Paithne Ajay N.

Abstract:

This paper proposes Intonation Modeling for Prosody generation in Neutral speech for Marathi (language spoken in Maharashtra, India) story telling applications. Nowadays audio story telling devices are very eminent for children. In this paper, we proposed tilt model for stressed words in Marathi for speech modification. Tilt model predicts modification in tone of neutral speech. GMM is used to identify stressed words for modification.

Keywords: tilt model, fundamental frequency, statistical parametric speech synthesis, GMM

Procedia PDF Downloads 393

Authors: Gautam Sharda

Abstract:

The concept of right speech is the third stage of the noble eightfold path as prescribed by the Buddha and followed by millions of practicing Buddhists. The Buddha lays a lot of importance on the notion of right speech (Samma Vacca). In the Angutara Nikaya, the Buddha mentioned what constitutes right speech, which is basically four kinds of abstentions; namely abstaining from false speech, abstaining from slanderous speech, abstaining from harsh or hateful speech and abstaining from idle chatter. The Buddha gives reasons in support of his view as to why abstaining from these four kinds of speeches is favourable not only for maintaining the peace and equanimity within an individual but also within a society. It is a known fact that when we say something harsh or slanderous to others, it eventually affects our individual peace of mind too. We also know about the many examples of hate speeches which have led to senseless cases of violence and which are well documented within our country and the world. Also, indulging in false speech is not a healthy sign for individuals within a group as this kind of a social group which is based on falsities and lies cannot really survive for long and will eventually lead to chaos. Buddha also told us to refrain from idle chatter or gossip as generally we have seen that idle chatter or gossip does more harm than any good to the individual and the society. Hence, if most of us actually inculcate this third stage (namely, right speech) of the noble eightfold path of the Buddha in our daily life, it would be highly beneficial both for the individual and for the harmony of the society.

Keywords: Buddhism, speech, individual, society

Procedia PDF Downloads 266

Authors: Jawharah Alasmari

Abstract:

In inflectional morphology of verb, tense and aspect indicate action’s time either past/present or future and their period whether completed or not. The usage and meaning of tense and aspect differ in Arabic and English, therefore is no simple one -to- one mapping from an Arabic verb inflected form an appropriate English translation depends on a range of features, including immediate and wider context of use. The Quranic Arabic Corpus includes seven alternative expertly crafted English translations of each Arabic verses, which provides a test dataset for the study of appropriate Arabic to English translations of verb tense and aspect. We applied Corpus Linguistics Methods in a theoretical study of exemplary verbs, to elicit candidate verbal contexts which influence the choice of English inflection for each verse.

Keywords: Corpus linguistics methods, Arabic verb, tense and aspect, English translations

Procedia PDF Downloads 392

Authors: Zelalem Fantahun

Abstract:

Part-of-speech tagging is the process of assigning a part-of-speech or other lexical class marker to each word into naturally occurring text. Part-of-speech tagging is the most fundamental and basic task almost in all natural language processing. In natural language processing, the problem of providing large amount of manually annotated data is a knowledge acquisition bottleneck. Since, Amharic is one of under-resourced language, the availability of tagged corpus is the bottleneck problem for natural language processing especially for POS tagging. A promising direction to tackle this problem is to provide a system that does not require manually tagged data. In unsupervised learning, the learner is not provided with classifications. Unsupervised algorithms seek out similarity between pieces of data in order to determine whether they can be characterized as forming a group. This paper explicates the development of unsupervised part-of-speech tagger using K-Means clustering for Amharic language since large amount of data is produced in day-to-day activities. In the development of the tagger, the following procedures are followed. First, the unlabeled data (raw text) is divided into 10 folds and tokenization phase takes place; at this level, the raw text is chunked at sentence level and then into words. The second phase is feature extraction which includes word frequency, syntactic and morphological features of a word. The third phase is clustering. Among different clustering algorithms, K-means is selected and implemented in this study that brings group of similar words together. The fourth phase is mapping, which deals with looking at each cluster carefully and the most common tag is assigned to a group. This study finds out two features that are capable of distinguishing one part-of-speech from others these are morphological feature and positional information and show that it is possible to use unsupervised learning for Amharic POS tagging. In order to increase performance of the unsupervised part-of-speech tagger, there is a need to incorporate other features that are not included in this study, such as semantic related information. Finally, based on experimental result, the performance of the system achieves a maximum of 81% accuracy.

Keywords: POS tagging, Amharic, unsupervised learning, k-means

Procedia PDF Downloads 452

Authors: Khaled M. Alhawiti

Abstract:

This research study aims to present a retrospective study about speech recognition systems and artificial intelligence. Speech recognition has become one of the widely used technologies, as it offers great opportunity to interact and communicate with automated machines. Precisely, it can be affirmed that speech recognition facilitates its users and helps them to perform their daily routine tasks, in a more convenient and effective manner. This research intends to present the illustration of recent technological advancements, which are associated with artificial intelligence. Recent researches have revealed the fact that speech recognition is found to be the utmost issue, which affects the decoding of speech. In order to overcome these issues, different statistical models were developed by the researchers. Some of the most prominent statistical models include acoustic model (AM), language model (LM), lexicon model, and hidden Markov models (HMM). The research will help in understanding all of these statistical models of speech recognition. Researchers have also formulated different decoding methods, which are being utilized for realistic decoding tasks and constrained artificial languages. These decoding methods include pattern recognition, acoustic phonetic, and artificial intelligence. It has been recognized that artificial intelligence is the most efficient and reliable methods, which are being used in speech recognition.

Keywords: speech recognition, acoustic phonetic, artificial intelligence, hidden markov models (HMM), statistical models of speech recognition, human machine performance

Procedia PDF Downloads 478

Authors: Vandana Mishra, Niladri Sekhar Dash, Jayshree Charkraborty

Abstract:

This present paper focuses on the application of corpus linguistics techniques for critical discourse analysis (CDA) of Hindi newspapers. While Corpus linguistics is the study of language as expressed in corpora (samples) of 'real world' text, CDA is an interdisciplinary approach to the study of discourse that views language as a form of social practice. CDA has mainly been studied from a qualitative perspective. However, we can say that recent studies have begun combining corpus linguistics with CDA in analyzing large volumes of text for the study of existing power relations in society. The corpus under our study is also of a sizable amount (1 million words of Hindi newspaper texts) and its analysis requires an alternative analytical procedure. So, we have combined both the quantitative approach i.e. the use of corpus techniques with CDA’s traditional qualitative analysis. In this context, we have focused on the Keyword Analysis Sorting Concordance Lines of the selected Keywords and calculating collocates of the keywords. We have made use of the Wordsmith Tool for all these analysis. The analysis starts with identifying the keywords in the political news corpus when compared with the main news corpus. The keywords are extracted from the corpus based on their keyness calculated through statistical tests like chi-squared test and log-likelihood test on the frequent words of the corpus. Some of the top occurring keywords are मोदी (Modi), भाजपा (BJP), कांग्रेस (Congress), सरकार (Government) and पार्टी (Political party). This is followed by the concordance analysis of these keywords which generates thousands of lines but we have to select few lines and examine them based on our objective. We have also calculated the collocates of the keywords based on their Mutual Information (MI) score. Both concordance and collocation help to identify lexical patterns in the political texts. Finally, all these quantitative results derived from the corpus techniques will be subjectively interpreted in accordance to the CDA’s theory to examine the ways in which political news discourse produces social and political inequality, power abuse or domination.

Keywords: critical discourse analysis, corpus linguistics, Hindi newspapers, power relations

Procedia PDF Downloads 225

Authors: Christian Arcos, Marley Vellasco, Abraham Alcaim

Abstract:

In this paper, we present a wavelet coefficients masking based on Local Binary Patterns (WLBP) approach to enhance the temporal spectra of the wavelet coefficients for speech enhancement. This technique exploits the wavelet denoising scheme, which splits the degraded speech into pyramidal subband components and extracts frequency information without losing temporal information. Speech enhancement in each high-frequency subband is performed by binary labels through the local binary pattern masking that encodes the ratio between the original value of each coefficient and the values of the neighbour coefficients. This approach enhances the high-frequency spectra of the wavelet transform instead of eliminating them through a threshold. A comparative analysis is carried out with conventional speech enhancement algorithms, demonstrating that the proposed technique achieves significant improvements in terms of PESQ, an international recommendation of objective measure for estimating subjective speech quality. Informal listening tests also show that the proposed method in an acoustic context improves the quality of speech, avoiding the annoying musical noise present in other speech enhancement techniques. Experimental results obtained with a DNN based speech recognizer in noisy environments corroborate the superiority of the proposed scheme in the robust speech recognition scenario.

Keywords: binary labels, local binary patterns, mask, wavelet coefficients, speech enhancement, speech recognition

Procedia PDF Downloads 230

Authors: Chafik Barnoussi, Mourad Talbi, Adnane Cherif

Abstract:

In this paper we propose a new speech compression system based on the application of the Bionic Wavelet Transform (BWT) combined with the psychoacoustic model. This compression system is a modified version of the compression system using a MDCT (Modified Discrete Cosine Transform) filter banks of 32 filters each and the psychoacoustic model. This modification consists in replacing the banks of the MDCT filter banks by the bionic wavelet coefficients which are obtained from the application of the BWT to the speech signal to be compressed. These two methods are evaluated and compared with each other by computing bits before and bits after compression. They are tested on different speech signals and the obtained simulation results show that the proposed technique outperforms the second technique and this in term of compressed file size. In term of SNR, PSNR and NRMSE, the outputs speech signals of the proposed compression system are with acceptable quality. In term of PESQ and speech signal intelligibility, the proposed speech compression technique permits to obtain reconstructed speech signals with good quality.

Keywords: speech compression, bionic wavelet transform, filterbanks, psychoacoustic model

Procedia PDF Downloads 384

Authors: Oxana Lukoshus

Abstract:

The methods of corpus linguistics, a well-established field of research, are being increasingly applied in cognitive linguistics. Corpora data are especially useful for different quantitative studies of grammatical and other aspects of language. The main objective of this paper is to demonstrate how present-day corpora can be applied in semantic studies in general and in semantic studies of adjectives in particular. Polysemantic adjectives have been the subject of numerous studies. But most of them have been carried out on dictionaries. Undoubtedly, dictionaries are viewed as one of the basic data sources, but only at the initial steps of a research. The author usually starts with the analysis of the lexicographic data after which s/he comes up with a hypothesis. In the research conducted three polysemantic synonyms true, loyal, faithful have been analyzed in terms of differences and similarities in their semantic structure. A corpus-based approach in the study of the above-mentioned adjectives involves the following. After the analysis of the dictionary data there was the reference to the following corpora to study the distributional patterns of the words under study – the British National Corpus (BNC) and the Corpus of Contemporary American English (COCA). These corpora are continually updated and contain thousands of examples of the words under research which make them a useful and convenient data source. For the purpose of this study there were no special needs regarding genre, mode or time of the texts included in the corpora. Out of the range of possibilities offered by corpus-analysis software (e.g. word lists, statistics of word frequencies, etc.), the most useful tool for the semantic analysis was the extracting a list of co-occurrence for the given search words. Searching by lemmas, e.g. true, true to, and grouping the results by lemmas have proved to be the most efficient corpora feature for the adjectives under the study. Following the search process, the corpora provided a list of co-occurrences, which were then to be analyzed and classified. Not every co-occurrence was relevant for the analysis. For example, the phrases like An enormous sense of responsibility to protect the minds and hearts of the faithful from incursions by the state was perceived to be the basic duty of the church leaders or ‘True,’ said Phoebe, ‘but I'd probably get to be a Union Official immediately were left out as in the first example the faithful is a substantivized adjective and in the second example true is used alone with no other parts of speech. The subsequent analysis of the corpora data gave the grounds for the distribution groups of the adjectives under the study which were then investigated with the help of a semantic experiment. To sum it up, the corpora-based approach has proved to be a powerful, reliable and convenient tool to get the data for the further semantic study.

Keywords: corpora, corpus-based approach, polysemantic adjectives, semantic studies

Procedia PDF Downloads 315

Authors: Theng Theng Ong

Abstract:

This pilot study adopts a corpus-based discourse analysis to explore the construction of Malaysia airline tragedy MH370 in the selected Malaysian and United Kingdom (UK) newspapers. Fairclough’s three-dimensional model is adopted in the study to support the corpus-based analysis. The analysis aims to determine the ways in which Malaysian Airline tragedy MH370 is linguistically defined and constructed in terms of keywords and collocation. The study also seeks to identify the types of discourse that are presented in the news articles. In addition, the differences or similarities in terms of keywords, topics or issues covered by the selected Malaysian and UK news media are examined.

Keywords: corpus, CDA, newspapers, airline tragedies

Procedia PDF Downloads 300

Authors: Nabil Shawkat, Jamil Saquer

Abstract:

Social media has accelerated our ability to engage with others and eliminated many communication barriers. On the other hand, the widespread use of social media resulted in an increase in online hate speech. This has drastic impacts on vulnerable individuals and societies. Therefore, it is critical to detect hate speech to prevent innocent users and vulnerable communities from becoming victims of hate speech. We investigate the performance of different deep learning and machine learning algorithms on three different datasets. Our results show that the BERT model gives the best performance among all the models by achieving an F1-score of 90.6% on one of the datasets and F1-scores of 89.7% and 88.2% on the other two datasets.

Keywords: hate speech, machine learning, deep learning, abusive words, social media, text classification

Procedia PDF Downloads 139

Authors: Samba Raju, Chiluveru, Manoj Tripathy

Abstract:

Intelligibility is an essential characteristic of a speech signal, which is used to help in the understanding of information in speech signal. Background noise in the environment can deteriorate the intelligibility of a recorded speech. In this paper, we presented a simple variance subtracted - variable level discrete wavelet transform, which improve the intelligibility of speech. The proposed algorithm does not require an explicit estimation of noise, i.e., prior knowledge of the noise; hence, it is easy to implement, and it reduces the computational burden. The proposed algorithm decides a separate decomposition level for each frame based on signal dominant and dominant noise criteria. The performance of the proposed algorithm is evaluated with speech intelligibility measure (STOI), and results obtained are compared with Universal Discrete Wavelet Transform (DWT) thresholding and Minimum Mean Square Error (MMSE) methods. The experimental results revealed that the proposed scheme outperformed competing methods

Keywords: discrete wavelet transform, speech intelligibility, STOI, standard deviation

Procedia PDF Downloads 149

Authors: Sulistyaningtyas

Abstract:

Examining the role of Middle Eastern freedom activists’ speech based on gender perspective is considered noteworthy because the society in the Middle East is patriarchal. This research aims to examine the language use of the Middle Eastern freedom activists’ speeches through gender perspective. The data sources are from male and female Middle Eastern freedom activists’ speech videos. In analyzing the data, the theories employed are about Language Style from Gender Perspective and The Language for Speech. The result reveals that there are sets of spoken language differences between male and female speakers. In using the language for speech, both male and female speakers produce metaphor, euphemism, the ‘rule of three’, parallelism, and pronouns in random frequency of production, which cannot be separated by genders. Moreover, it cannot be concluded that one gender is more potential than the other to influence the audience in delivering speech. There are other factors, particularly non-verbal factors, existing to give impacts on how a speech can influence the audience.

Keywords: gender perspective, language use, Middle Eastern freedom activists, speech

Procedia PDF Downloads 423

Authors: Simona Maria Ignat

Abstract:

Bullying online has been an increasing studied topic during the last years. Different approaches, psychological, linguistic, or computational, have been applied. To our best knowledge, a definition and a set of characteristics of phenomenon agreed internationally as a common framework are still waiting for answers. Thus, the objectives of this paper are the identification of bullying utterances on Twitter and their algorithms. This research paper is focused on the identification of words or groups of words, categorized as “utterances”, with bullying effect, from Twitter platform, extracted on a set of legislative criteria. This set is the result of analysis followed by synthesis of law documents on bullying(online) from United States of America, European Union, and Ireland. The outcome is a linguistic corpus with approximatively 10,000 entries. The methods applied to the first objective have been the following. The discourse analysis has been applied in identification of keywords with bullying effect in texts from Google search engine, Images link. Transcription and anonymization have been applied on texts grouped in CL1 (Corpus linguistics 1). The keywords search method and the legislative criteria have been used for identifying bullying utterances from Twitter. The texts with at least 30 representations on Twitter have been grouped. They form the second corpus linguistics, Bullying utterances from Twitter (CL2). The entries have been identified by using the legislative criteria on the the BoW method principle. The BoW is a method of extracting words or group of words with same meaning in any context. The methods applied for reaching the second objective is the conversion of parts of speech to alphabetical and numerical symbols and writing the bullying utterances as algorithms. The converted form of parts of speech has been chosen on the criterion of relevance within bullying message. The inductive reasoning approach has been applied in sampling and identifying the algorithms. The results are groups with interchangeable elements. The outcomes convey two aspects of bullying: the form and the content or meaning. The form conveys the intentional intimidation against somebody, expressed at the level of texts by grammatical and lexical marks. This outcome has applicability in the forensic linguistics for establishing the intentionality of an action. Another outcome of form is a complex of graphemic variations essential in detecting harmful texts online. This research enriches the lexicon already known on the topic. The second aspect, the content, revealed the topics like threat, harassment, assault, or suicide. They are subcategories of a broader harmful content which is a constant concern for task forces and legislators at national and international levels. These topic – outcomes of the dataset are a valuable source of detection. The analysis of content revealed algorithms and lexicons which could be applied to other harmful contents. A third outcome of content are the conveyances of Stylistics, which is a rich source of discourse analysis of social media platforms. In conclusion, this corpus linguistics is structured on legislative criteria and could be used in various fields.

Keywords: corpus linguistics, cyberbullying, legislation, natural language processing, twitter

Procedia PDF Downloads 86

Authors: Zdeněk Joukl

Abstract:

The word for pig in Czech (prase) seems to be one of the most negatively connotated words denoting animals. This paper represents an analysis of the largest Czech corpora, including a diachronic corpus. Besides corpus-analytical tools, sentiment analysis methods and tools such as LIWC and word clouds are used to better capture the usage of the words for pigs in Czech. The most frequent collocations across domains are identified and extracted with context to be used for sentiment analysis, which reveals an almost exclusive negative sentiment or culinary context. The animal is burdened with a disproportionately high number of meanings representing negatively viewed human characteristics or behaviors (dirtiness, fatness, sweating, inebriation, aggressive driving, greediness or chauvinism are among the most frequent ones). The diachronic view helps us understand how this extreme bias came to existence both through institutional construction and human-animal relations.

Keywords: corpus analysis, pig, sentiment, social construction

Procedia PDF Downloads 14

Authors: Gabriela Smeckova

Abstract:

The entire world is becoming more and more diverse. The reasons why people migrate are different and unique for each family /individual. Professionals delivering services (including speech-language pathologists) must be prepared to work with clients coming from different cultural and/or linguistic backgrounds. Well-educated speech-language pathologists will consider many factors when delivering services. Some of them will be discussed during the presentation (language spoken, beliefs about health care and disabilities, reasons for immigration, etc.). The communication styles of the client can be different than the styles of the speech-language pathologist. The goal is to become culturally responsive in service delivery.

Keywords: culture, cultural competence, culturallly responsive practices, speech-language pathologist, cultural and linguistical variables, communication styles

Procedia PDF Downloads 78

Authors: Sham Kidane, Fitsum Gaim, Ibrahim Abdella, Sirak Asmerom, Yoel Ghebrihiwot, Simon Mulugeta, Natnael Ambassager

Abstract:

The lack of annotated datasets is a bottleneck to the progress of NLP in low-resourced languages. The work presented here consists of large-scale annotated datasets and models for the named entity recognition (NER) system for the Tigrinya language. Our manually constructed corpus comprises over 340K words tagged for NER, with over 118K of the tokens also having parts-of-speech (POS) tags, annotated with 12 distinct classes of entities, represented using several types of tagging schemes. We conducted extensive experiments covering convolutional neural networks and transformer models; the highest performance achieved is 88.8% weighted F1-score. These results are especially noteworthy given the unique challenges posed by Tigrinya’s distinct grammatical structure and complex word morphologies. The system can be an essential building block for the advancement of NLP systems in Tigrinya and other related low-resourced languages and serve as a bridge for cross-referencing against higher-resourced languages.

Keywords: Tigrinya NER corpus, TiBERT, TiRoBERTa, BiLSTM-CRF

Procedia PDF Downloads 133

Authors: Rajani S. Pujar, Pandurangarao N. Kulkarni

Abstract:

Increased temporal masking affects the speech perception in persons with sensorineural hearing impairment especially under adverse listening conditions. This paper presents a cascaded scheme, which employs a noise reduction algorithm as well as temporal splitting of the speech signal. Earlier investigations have shown that by splitting the speech temporally and presenting alternate segments to the two ears help in reducing the effect of temporal masking. In this technique, the speech signal is processed by two fading functions, complementary to each other, and presented to left and right ears for binaural dichotic presentation. In the present study, half cosine signal is used as a fading function with crossover gain of 6 dB for the perceptual balance of loudness. Temporal splitting is combined with noise reduction algorithm to improve speech perception in the background noise. Two noise reduction schemes, namely spectral subtraction and Wiener filter are used. Listening tests were conducted on six normal-hearing subjects, with sensorineural loss simulated by adding broadband noise to the speech signal at different signal-to-noise ratios (∞, 3, 0, and -3 dB). Objective evaluation using PESQ was also carried out. The MOS score for VCV syllable /asha/ for SNR values of ∞, 3, 0, and -3 dB were 5, 4.46, 4.4 and 4.05 respectively, while the corresponding MOS scores for unprocessed speech were 5, 1.2, 0.9 and 0.65, indicating significant improvement in the perceived speech quality for the proposed scheme compared to the unprocessed speech.

Keywords: MOS, PESQ, spectral subtraction, temporal splitting, wiener filter

Procedia PDF Downloads 328

Authors: Ajish K. Abraham

Abstract:

Hearing aid is the most fundamental technology employed towards rehabilitation of persons with sensory neural hearing impairment. Hearing in noise is still a matter of major concern for many hearing aid users and thus continues to be a challenging issue for the hearing aid designers. Several techniques are being currently used to enhance the speech at the hearing aid output. Most of these techniques, when implemented, result in reduction of intelligibility of the speech signal. Thus the dissatisfaction of the hearing aid user towards comprehending the desired speech amidst noise is prevailing. Multichannel Wiener Filter is widely implemented in binaural hearing aid technology for noise reduction. In this study, Wiener filter based noise reduction approach is experimented for a single microphone based hearing aid set up. This method checks the status of the input speech signal in each frequency band and then selects the relevant noise reduction procedure. Results showed that the Wiener filter based algorithm is capable of enhancing speech even when the input acoustic signal has a very low Signal to Noise Ratio (SNR). Performance of the algorithm was compared with other similar algorithms on the basis of improvement in intelligibility and SNR of the output, at different SNR levels of the input speech. Wiener filter based algorithm provided significant improvement in SNR and intelligibility compared to other techniques.

Keywords: hearing aid output speech, noise reduction, SNR improvement, Wiener filter, speech enhancement

Procedia PDF Downloads 247

Authors: Mumtaz Begum Mustafa, Siti Salwah Salim, Feizal Dani Rahman

Abstract:

Recently, Automatic Speech Recognition (ASR) systems were used to assist children in language acquisition as it has the ability to detect human speech signal. Despite the benefits offered by the ASR system, there is a lack of ASR systems for Malay-speaking children. One of the contributing factors for this is the lack of continuous speech database for the target users. Though cross-lingual adaptation is a common solution for developing ASR systems for under-resourced language, it is not viable for children as there are very limited speech databases as a source model. In this research, we propose a two-stage adaptation for the development of ASR system for Malay-speaking children using a very limited database. The two stage adaptation comprises the cross-lingual adaptation (first stage) and cross-age adaptation. For the first stage, a well-known speech database that is phonetically rich and balanced, is adapted to the medium-sized Malay adults using supervised MLLR. The second stage adaptation uses the speech acoustic model generated from the first adaptation, and the target database is a small-sized database of the target users. We have measured the performance of the proposed technique using word error rate, and then compare them with the conventional benchmark adaptation. The two stage adaptation proposed in this research has better recognition accuracy as compared to the benchmark adaptation in recognizing children’s speech.

Keywords: Automatic Speech Recognition System, children speech, adaptation, Malay

Procedia PDF Downloads 398

Authors: Emma Nemishalyan

Abstract:

It is believed that learners’ mother tongue (L1 hereafter) has a huge impact on their second language acquisition (L2 hereafter). This hypothesis has been exposed to both positive and negative criticism. Based on research results of a wide range of learners’ corpora (Chinese, Japanese, Spanish among others) the hypothesis has either been proved or disproved. However, no such study has been conducted on the Armenian learners. The aim of this paper is to understand the implication of the hypothesis on the Armenian learners’ corpus in terms of the use of the passive voice. To this end, the method of Contrastive Interlanguage Analysis (hereafter CIA) has been used on native speakers’ corpus (Louvain Corpus of Native English Essays (LOCNESS)) and Armenian learners’ corpus which has been compiled by me in compliance with International Corpus of Learner English (ICLE) guidelines. CIA compares the interlanguage (the language produced by learners) with the one produced by native speakers. With the help of this method, it is possible not only to highlight the mistakes that learners make, but also to underline the under or overuses. The choice of the grammar issue (passive voice) is conditioned by the fact that typologically Armenian and English are drastically different as they belong to different branches. Moreover, the passive voice is considered to be one of the most problematic grammar topics to be acquired by learners of the English language. Based on this difference, we hypothesized that Armenian learners would either overuse or underuse some types of the passive voice. With the help of Lancsbox software, we have identified the frequency rates of passive voice usage in LOCNESS and Armenian learners’ corpus to understand whether the latter have the same usage pattern of the passive voice as the native speakers. Secondly, we have identified the types of the passive voice used by the Armenian leaners trying to track down the reasons in their mother tongue. The results of the study showed that Armenian learners underused the passive voices in contrast to native speakers. Furthermore, the hypothesis that learners’ L1 has an impact on learners’ L2 acquisition and production was proved.

Keywords: corpus linguistics, applied linguistics, second language acquisition, corpus compilation

Procedia PDF Downloads 110

Authors: Tanju Deveci

Abstract:

It appears that the speech act of complaint has not received as much attention as other speech acts. However, the face-threatening nature of this speech act requires a special attention in multicultural contexts in particular. The teaching context in the UAE universities, where a big majority of teaching staff comes from other cultures, requires investigations into this speech act in order to improve communication between students and faculty. This session will outline the results of a study conducted with this purpose. The realization of complaints by Freshman English students in Communication courses at Petroleum Institute was investigated to identify communication patterns that seem to cause a strain. Data were collected using a role-play between a teacher and students, and a judgment scale completed by two of the instructors in the Communications Department. The initial findings reveal that the students had difficulty putting their case, produced the speech act of criticism along with a complaint and that they produced both requests and demands as candidate solutions. The judgement scales revealed that the students’ attitude was not appropriate most of the time and that the judges would behave differently from students. It is concluded that speech acts, in general, and complaint, in particular, need to be taught to learners explicitly to improve interpersonal communication in multicultural societies. Some teaching ideas are provided to help increase foreign language learners’ sociolinguistic competence.

Keywords: speech act, complaint, pragmatics, sociolinguistics, language teaching

Procedia PDF Downloads 509

Authors: Jian-Jun Gao

Abstract:

Repetition is used for a variety of functions in conversation. When young children first learn to speak, they often repeat words from the adult’s recent utterance with the learning and social function. The objective of this study was to ascertain whether the repetitions are equivalent in indicating attention to new words and the initial repeat of information in conversation. Based on the observation of naturally occurring language use in Taiwan Corpus of Child Mandarin (TCCM), the results in this study provided empirical support to the previous findings that children are more likely to repeat new words they are offered than to repeat new information. When children get older, there would be a drop in the repetition of both new words and new information.

Keywords: acquisition, corpus, mandarin, new words, new information, repetition

Procedia PDF Downloads 149

Authors: Jingwen Ou

Abstract:

Writing for academic purposes in a second or foreign language poses a significant challenge for non-native speakers, particularly at the tertiary level, where English academic writing for L2 students is often hindered by difficulties in academic discourse, including vocabulary, academic register, and organization. The past two decades have witnessed a rising popularity in the application of the data-driven learning (DDL) approach in EAP writing instruction. In light of such a trend, this study aims to enhance the integration of DDL into English for academic purposes (EAP) writing classrooms by investigating the perception of Chinese college students regarding the use of corpus tools for improving EAP writing. Additionally, the research explores their corpus consultation behaviors during training to provide insights into corpus-assisted EAP instruction for DDL practitioners. Given the uprising popularity of DDL, this research aims to investigate Chinese university students’ use of corpus tools with three main foci: 1) the influence of corpus tools on learning behaviours, 2) the influence of corpus tools on students’ academic writing performance outcomes, and 3) students’ perceptions and potential perceptional changes towards the use of such tools. Three corpus tools, CQPWeb, Sketch Engine, and LancsBox X, are selected for investigation due to the scarcity of empirical research on patterns of learners’ engagement with a combination of multiple corpora. The research adopts a pre-test / post-test design for the evaluation of students’ academic writing performance before and after the intervention. Twenty participants will be divided into two groups: an intervention and a non-intervention group. Three corpus training workshops will be delivered at the beginning, middle, and end of a semester. An online survey and three separate focus group interviews are designed to investigate students’ perceptions of the use of corpus tools for improving academic writing skills, particularly the rhetorical functions in different essay sections. Insights from students’ consultation sessions indicated difficulties with DDL practice, including insufficiency of time to complete all tasks, struggle with technical set-up, unfamiliarity with the DDL approach and difficulty with some advanced corpus functions. Findings from the main study aim to provide pedagogical insights and training resources for EAP practitioners and learners.

Keywords: corpus linguistics, data-driven learning, English for academic purposes, tertiary education in China

Procedia PDF Downloads 62