Speech Transcription Research Articles

THE REPRESENTATION OF IRAN AND UNITED STATES IN DONALD TRUMP’S SPEECH: A CRITICAL DISCOURSE ANALYSIS

Speech plays an important role to shape public perception as it is delivered by an influential figure and reflects the points of view of its speaker. This study discusses a speech delivered by the United States President, Donald Trump, which discusses the conflict between Iran and America. This study was conducted to reveal the representation of Iran and the United States in one of Donald Trump’s speeches. The method used in this study is the descriptive qualitative method. The data in this study are in the form of words, phrases, clauses, and sentences that indicate the position of Iran and the United States taken from the speech transcript from the official website of The White House. Data were analyzed under the framework of Van Djik’s Socio-Cognitive Approach, consisting of text, socio-cognitive, and social context. Based on the analysis, it can be concluded that Iran is described as a country that has nuclear ambitions and acts of terror that support the existence of terrorists. On the other hand, America is described as having an invincible power. At a socio-cognitive level, Donald Trump is considered a knowledgeable figure on his country’s political condition because he knows the weaknesses of Iran and can properly take every decision. Donald Trump also has the authority as a President to make The United States and the countries of the world can work together for peace world. Thus, from the level of social context, countries in the world support the actions taken by America and are very alarming about what Iran has done. The finding suggests that the Socio-cognitive approach is practical to analyze the representation of an issue in speech reflected in linguistics expression and the discourse structure.

Comparative Study of Generating Metacognitive Competence Strategy of Elementary School Students

This study aims to compare teachers' strategy in generating students’ metacognitive abilities in elementary schools. The substantive data source of this research is the texts produced by transcription of teacher's speech records in carrying out the teaching-learning process with students in which there are strategies for generating students' metacognitive abilities. The locative data sources of this research are Demangan Baru Kanisius Elementary School Yogyakarta (Elementary School A) and Pius Bhakti Utama Elementary School Kutoarjo (Elementary School B). The data of this study were the form of speech footage from the transcription of teacher records containing strategies for generating students' metacognition. The data collection method used was the method of listening, both referring to both involved and able to see and be involved freely. The technique used was the note taking technique and record technique. Furthermore, the data analysis method used was the content analysis method. The results of the study were presented by the informal presentation method. The results indicated that the metacognitive competence generation strategies undertaken by teachers in School A and School B include the following strategies: (1) affirm critical thinkers' self-worth, (2) listen attentively to critical thinkers, (3) show that you support critical thinkers 'efforts, (4) reflect and mirror critical thinkers' ideas and actions, (5) motivate people to think critically, (6) regularly evaluate progress, (7) help critical thinkers create networks, (8) be critical teachers . The strategy of making people aware of how they learn critical thinking is only implemented in one school. The realization of students’ metacognitive competence strategies in the two schools varies according to the experience of the teacher in teaching, the background of students' abilities, and the perspective of the teachers in learning, etc.

Interpersonal functions in Greta Thunberg’s “civil society for rEUnaissance” speech

Greta Thunberg delivered a speech entitled “You’re Acting Like Spoiled, Irresponsible Children” to influential figures in Europe through the “Civil Society for renaissance” event, in which she was personally invited by the president of the organizer, Luca Jahier. Through her speech, she managed to convince the European Union to pledge to spend billions of euros to combat the climate crisis. This study aims to discover how interpersonal metafunction is used in the speech and what functions are revealed through Fairclough’s Critical Discourse Analysis and Hallidayan Systemic Functional Grammar. The data, consisting of seventy-one independent clauses, were taken from Thunberg’s book which consists of her speech transcription, titled “No One Is Too Small to Make a Difference”. The research discovers that through the use of mood, modality, and pronouns, Thunberg conveys her view on her relationship with the audience as victim and perpetrator, and the one who holds responsibility and takes the blame.

Evandro, menino desatento? Mal-estar e medicalização na educação

This article is the result of an intervention research based on a case study. It carries out aposteriori analysis of the material referring to the school discourse about a child undergoing schooling,aiming to understand the production of malaise. Based on reports of visits to the school, school reportsand transcripts of speeches by school agents, it reflects on the malaise in the educators’ discoursetowards the student who does not correspond to the ideal, as well as maps the strategies to overcomethe malaise. The article concludes that, in the case studied, the strategies predominantly pointed towhat can be called the medicalization of education in the contemporary.

Consistent Transcription and Translation of Speech

The conventional paradigm in speech translation starts with a speech recognition step to generate transcripts, followed by a translation step with the automatic transcripts as input. To address various shortcomings of this paradigm, recent work explores end-to-end trainable direct models that translate without transcribing. However, transcripts can be an indispensable output in practical applications, which often display transcripts alongside the translations to users. We make this common requirement explicit and explore the task of jointly transcribing and translating speech. Although high accuracy of transcript and translation are crucial, even highly accurate systems can suffer from inconsistencies between both outputs that degrade the user experience. We introduce a methodology to evaluate consistency and compare several modeling approaches, including the traditional cascaded approach and end-to-end models. We find that direct models are poorly suited to the joint transcription/translation task, but that end-to-end models that feature a coupled inference procedure are able to achieve strong consistency. We further introduce simple techniques for directly optimizing for consistency, and analyze the resulting trade-offs between consistency, transcription accuracy, and translation accuracy. 1

Impact of Trump's Promotion of Unproven COVID-19 Treatments and Subsequent Internet Trends: Observational Study.

BackgroundIndividuals with large followings can influence public opinions and behaviors, especially during a pandemic. In the early days of the pandemic, US president Donald J Trump has endorsed the use of unproven therapies. Subsequently, a death attributed to the wrongful ingestion of a chloroquine-containing compound occurred.ObjectiveWe investigated Donald J Trump’s speeches and Twitter posts, as well as Google searches and Amazon purchases, and television airtime for mentions of hydroxychloroquine, chloroquine, azithromycin, and remdesivir.MethodsTwitter sourcing was catalogued with Factba.se, and analytics data, both past and present, were analyzed with Tweet Binder to assess average analytics data on key metrics. Donald J Trump’s time spent discussing unverified treatments on the United States’ 5 largest TV stations was catalogued with the Global Database of Events, Language, and Tone, and his speech transcripts were obtained from White House briefings. Google searches and shopping trends were analyzed with Google Trends. Amazon purchases were assessed using Helium 10 software.ResultsFrom March 1 to April 30, 2020, Donald J Trump made 11 tweets about unproven therapies and mentioned these therapies 65 times in White House briefings, especially touting hydroxychloroquine and chloroquine. These tweets had an impression reach of 300% above Donald J Trump’s average. Following these tweets, at least 2% of airtime on conservative networks for treatment modalities like azithromycin and continuous mentions of such treatments were observed on stations like Fox News. Google searches and purchases increased following his first press conference on March 19, 2020, and increased again following his tweets on March 21, 2020. The same is true for medications on Amazon, with purchases for medicine substitutes, such as hydroxychloroquine, increasing by 200%.ConclusionsIndividuals in positions of power can sway public purchasing, resulting in undesired effects when the individuals’ claims are unverified. Public health officials must work to dissuade the use of unproven treatments for COVID-19.

Depression-level assessment from multi-lingual conversational speech data using acoustic and text features

Depression is a widespread mental health problem around the world with a significant burden on economies. Its early diagnosis and treatment are critical to reduce the costs and even save lives. One key aspect to achieve that goal is to use technology and monitor depression remotely and relatively inexpensively using automated agents. There has been numerous efforts to automatically assess depression levels using audiovisual features as well as text-analysis of conversational speech transcriptions. However, difficulty in data collection and the limited amounts of data available for research present challenges that are hampering the success of the algorithms. One of the two novel contributions in this paper is to exploit databases from multiple languages for acoustic feature selection. Since a large number of features can be extracted from speech, given the small amounts of training data available, effective data selection is critical for success. Our proposed multi-lingual method was effective at selecting better features than the baseline algorithms, which significantly improved the depression assessment accuracy. The second contribution of the paper is to extract text-based features for depression assessment and use a novel algorithm to fuse the text- and speech-based classifiers which further boosted the performance.

Resistir para/com a infância continuar a existir: Transcrição de 21 lives sobre infâncias e Educação Infantil na pandemia de Covid-19

Partindo da emergência de considerarmos as crianças e suas infâncias nas propostas voltadas à Educação Infantil, nesse cenário de educação não presencial no contexto da pandemia, nesta publicação da seção ‘Outras linguagens’, da revista Zero a Seis, reunimos transcrições de falas extraídas de vídeos ao vivo (lives) transmitidos pelo Youtube no período de abril a julho de 2020. As lives foram localizadas a partir de uma pesquisa prévia realizada pela pesquisadora Natalia Francisca Cardia dos Santos, publicada na edição especial “Covid-19: Educação em tempos de pandemia”, da Revista Olhar de professor, e selecionadas usando como critério tratarem de temáticas relacionadas à Educação Infantil em tempos de Covid-19, sob a ótica da defesa das crianças e das infâncias, com aporte na Pedagogia da infância e na Sociologia da infância. Avaliando ser essencial possibilitar uma visão interseccional da conjuntura social e política que permeia a educação e as políticas educacionais, incluímos também em nossa seleção algumas lives em que foram abordados temas como racismo na Educação Infantil, relações de gênero e a relação entre feminismo e políticas públicas de educação para a Educação Infantil, transmitidas até o mês de agosto do mesmo ano. Vivemos um tempo sem precedentes, em que temas importantes como estes estão sendo discutidos, estudados e refletidos conjuntamente por meio de telas de computador e compartilhadas em mídias sociais, que, embora se apresentem como uma estratégia legítima de mobilização, implicam risco de que percamos a preciosidade dessas trocas e análises em meio ao tempo e à lógica da velocidade digital. Ao nos lançarmos ao desafio de mergulhar em outras linguagens, juntamente com as(os) quarenta e nove pesquisadoras(es) que aceitaram o convite de se juntarem a nós, reunimos diálogos de vários lugares do Brasil, com registros transcritos pelas(os) próprias(os) autoras(es), fiéis do que foi dito naquele momento, salvo algumas alterações efetuadas em palavras necessárias para possibilitar a compreensão do que foi explicitado e cortes de trechos mais longos, e/ou exemplos tornando a estrutura da escrita/leitura mais objetiva e compacta, mas preservando o conteúdo, que permanecem com as marcas de oralidade e da contextualidade espaço-temporal de quando foram transmitidos ao vivo, e organizados por data da transmissão. Deste modo, ensejamos que as próprias falas transcritas deem conta de localizar os acontecimentos nessa delimitação cronológica, funcionando como uma espécie de retrospectiva do que foi discutido acerca de Educação Infantil no mencionado recorte de tempo. Temos certeza de que, ao conferirmos materialidade às lives, que somam tantos olhares e pontos de vista dessa temática, compusemos um acervo documental em múltiplos eixos, e daqui seguimos com a esperança de que a união da força da palavra dita à palavra escrita seja instrumento potente e relevante para quem arrisque, neste ou em outro tempo, sonhar e viver a educação em prol de uma vida mais cidadã para todas as pessoas, resistindo e ousando com a infância, para continuar a existir em tempos de Covid-19 e no que vier a partir desse período. “Queremos um mundo onde sejamos socialmente iguais, humanamente diferentes e totalmente livres” (Rosa de Luxemburgo)

Denial of the Rohingya Genocide: Problematising Daw Aung San Suu Kyi and the Rule of Law in Postcolonial Myanmar

As the world condemns the genocide, Myanmar and Daw Aung San Suu Kyi’s government denied it and refused to accept it and argued that it sanctioned under the rule of law. This paper problematises these questions:, What is the rule of law in Myanmar? And why do they deny it? This study is qualitative in which pages of transcripts of speeches perused to find themes, settings, and meanings attributed to problematising Daw Aung San Suu Kyi and the rule of law in post-colonial Myanmar. These speeches delivered in public from 2016-2018. In analysing her speeches, the paper uses Foucauldian Discourse Analysis. On the part of Daw Aung San Suu Kyi, it found that problematising her actions and silence over the Rohingya genocide influenced her late father’s role. She continued to claim that her military father is the father of the Burmese military. She had special relations in incarceration with the army generals during her house arrest. She focused on the democratic transition which she promised in the 2015 election. This study reveals that this rule of law has purely political narratives because the generals are not accountable. This paper subsumes ongoing legal reforms in Myanmar.

Perceptions of Artificial Intelligence Among Healthcare Staff: A Qualitative Survey Study.

Objectives: The medical community is in agreement that artificial intelligence (AI) will have a radical impact on patient care in the near future. The purpose of this study is to assess the awareness of AI technologies among health professionals and to investigate their perceptions toward AI applications in medicine. Design: A web-based Google Forms survey was distributed via the Royal Free London NHS Foundation Trust e-newsletter. Setting: Only staff working at the NHS Foundation Trust received an invitation to complete the online questionnaire. Participants: 98 healthcare professionals out of 7,538 (response rate 1.3%; CI 95%; margin of error 9.64%) completed the survey, including medical doctors, nurses, therapists, managers, and others. Primary outcome: To investigate the prior knowledge of health professionals on the subject of AI as well as their attitudes and worries about its current and future applications. Results: 64% of respondents reported never coming across applications of AI in their work and 87% did not know the difference between machine learning and deep learning, although 50% knew at least one of the two terms. Furthermore, only 5% stated using speech recognition or transcription applications on a daily basis, while 63% never utilize them. 80% of participants believed there may be serious privacy issues associated with the use of AI and 40% considered AI to be potentially even more dangerous than nuclear weapons. However, 79% also believed AI could be useful or extremely useful in their field of work and only 10% were worried AI will replace them at their job. Conclusions: Despite agreeing on the usefulness of AI in the medical field, most health professionals lack a full understanding of the principles of AI and are worried about potential consequences of its widespread use in clinical practice. The cooperation of healthcare workers is crucial for the integration of AI into clinical practice and without it the NHS may miss out on an exceptionally rewarding opportunity. This highlights the need for better education and clear regulatory frameworks.

Le sourire du nouveau-né sous le regard du parent

ObjectiveFrom the first hours after birth and from a morphological standpoint, newborns smile in a way similar to older children. Nevertheless, the newborn's smile has long been described as endogenous, thus implicitly assigning no social value to it. The objective of this exploratory study is to question the social nature of the newborn's smile during an interactive situation. Although this is a favorable context, it has seldom been explored inside the literature. Our hypothesis is that the social nature of the newborn's smile is linked to the exchange dynamic, where the parent reacts to the smile and gives meaning to it. MethodFifteen mother/2 to 4-days-old newborn dyads and 21 father/1 to 7-days-old newborn dyads were video-recorded at the maternity. Video coding was carried out frame by frame using the software The Observer XT. Transcription of parental speech was done using the software Audacity. ResultsOur results go hand in hand with our initial hypothesis. First of all, the parent smiles or continues to smile following the newborn's smile. Moreover, the parent's successive verbalizations evoke the newborn's smile in an unambiguous manner. In addition, the way that smile is described, one time out of three, suggests it is perceived as ‘directed’. ConclusionRather than lying on its voluntary or involuntary nature, the social dimension of the newborn's smile lies on the intimate feeling experienced by the partner, during the interaction, that the smile is directed towards him or her.

Release in linguistic masking changes over time: differences between native and non-native listeners

Research into speech-on-speech masking has identified what is known as release from linguistic masking, i.e., better speech transcription performance against a masker in an unknown than known language. To test whether native (English) and non-native (Mandarin) speakers of English could learn to control the interference of a known language masker, we measured their ability to transcribe English sentences against the background of English or Mandarin competing talkers over the course of 50 trials. Both groups improved over time. Native listeners exhibited release from linguistic masking, with less masking in the Mandarin than English masker condition. The size of this effect increased over time. In contrast, non-native listeners showed no difference between the two language maskers, with this pattern unaffected by time. Masker-to-target intrusion errors decreased over time for native listeners, whereas they were virtually absent for the non-native listeners. The results show that (1) Linguistic masking is worst when the masker language is known to the listener, whether that language is native or non-native, (2) Whether the masker is the same language as the target language is comparatively less important for non-native listeners, and (3) Native listeners are better at learning to suppress masker interference over time than non-native listeners.

Аргументативные маркеры в испанском политическом дискурсе

Статья посвящена изучению особенностей функционирования дискурсивных маркеров в испанском политическом дискурсе как эксплицитных средств актуализации аргументации в речи испанских политиков. Рассматриваются такие важные для целей исследования проблемы, как основные подходы к пониманию политического дискурса и к трактовке дискурсивных маркеров. Особое внимание уделяется классификации дискурсивных маркеров аргументации в зарубежной испанистике, описываются основные виды аргументативных маркеров и средства их репрезентации в испанском языке. В качестве материала исследования использовались выступления на заседаниях в Конгрессе депутатов Генеральных Кортесов Испании шести кандидатов на пост главы испанского правительства. Общий объем стенограмм публичных выступлений Леопольда Калво (1981), Фелипе Гонсалеса (1989), Марии Хосе Азнара (1996), Хосе Луиса Сапатеро (2004), Марьяно Рахоя (2016) и Педро Санчеса (2016) составляет 51 781 словоупотребление. Методом сплошной выборки в них было выделено 285 случаев употребления аргументативных маркеров. Для анализа языкового материала использовались метод наблюдения, сопоставительный метод, а также метод количественного анализа. В результате проведенного исследования были выявлены частотность и особенности функционирования разных видов аргументативных маркеров (экспликативов, консекутивов и контраргументативов) в испанском политическом дискурсе и в выступлениях отдельных политических деятелей. Были сделаны выводы о закономерностях использования языковых средств репрезентации отдельных аргументативных маркеров в публичных выступлениях политиков в одинаковых коммуникативно-дискурсивных условиях, о взаимосвязи использования разных аргументативных стратегий, приемов и тактик и языковой личности политика.

The conceptualization of morality in modern english-language globalization discourse: evidence from the United Nations Speeches

The increased interest of international cognitive linguists in the mechanisms of conceptualizing modern social phenomena has necessitated cognitive linguistic analysis of such phenomena as globalization, which is one of the most important trends setting the vector for modern society development.This study attempts not only to identify key concepts and means of their representation in terms of the globalization phenomenon, but also to build with the help of these concepts elements of the modern moral system inherent in the English-speaking community. To this end, a conceptual and cognitive-semantic analysis of contemporary English-language political discourse was carried out on the basis of speeches delivered by delegates to the United Nations.The investigation is premised on the theory of conceptual metaphor, emphasizing the need to understand the metaphorical foundations of human consciousness and communication. The study collected and analyzed empirical data that can be used to draw conclusions about the models of representation and assessment of reality by members of the English-speaking community, which in turn opens up prospects for further research in a linguistic pragmatic way and studying the specific features of English-speakers’ view of the world.As a result of lexicological and discourse-based analysis of speech transcripts, the paper uncovers several basic metaphorical models (Morality as 1. Commitment; 2. Nurturant Parent; 3. Resilience; 4. Fairy Tale of the Just War; 5. Progress), which outline globalization within the conceptual view of the world and which are underlined by such antitheses as “moral - immoral,” “success - loss,” “strength - weakness” etc.

Over-generalization, Under-generalization, and No-generalization on a Child’s Language Acquisition

It seems that many young children may not comprehend the word-reference relationship even though they can produce the words accurately. Since it happens, the children may face the situation known as over-generalization, under-generalization, and no-generalization. This study aims to get an in-depth understanding of over-generalization, under-generalization, and no-generalization on a child’s language acquisition. The method of research was a case study. The subject of research was a two-year-old child. Data of research were collected from the result of participant observation and documentation of speech transcription. Data analysis of this research covered data reduction, data display, and conclusion drawing or verification. The result shows that (1) over-generalization, under-generalization, and no-generalization occur when the subject aged 26 to 30 months; (2) at aged 32 months, those errors are finally revealed since the subject can manage his cognitive development and conceptualize his semantic-reference relationship for particular properties; and (3) this also proves that environment has a massive role to support and stimulate a child in acquiring and producing his language

Analytical review of approaches for tonality recognition of Russian text data

Introduction: In recent years, sentiment analysis has found practical application in many areas, such as evaluating the quality of products and services based on customers’ online reviews, analyzing negative emotions in messages, forecasting stock markets or political situations based on news data. In this regard, a large number of systems and methods for Russian text sentiment analysis are being developed. Purpose: A detailed review of approaches, and comparative analysis of available databases in the field of Russian text sentimental analysis. Results: Our analytical review of the approaches to Russian text data sentiment analysis has shown that there are a large number of ways for preprocessing, vectorization and machine classification of the text data. Studying the available databases shows that the Russian text sentimental analysis is less developed than that for other major world languages. Studying the existing software systems for Russian text analysis reveals their low accuracy compared to English, which can be caused by the sophisticated structure of Russian. Discussion: In our further research, we plan to implement sentiment analysis of spoken speech using audio data. To do this, we will need to obtain a spelling transcription of speech for each speaker.

Subs2vec: Word embeddings from subtitles in 55 languages

This paper introduces a novel collection of word embeddings, numerical representations of lexical semantics, in 55 languages, trained on a large corpus of pseudo-conversational speech transcriptions from television shows and movies. The embeddings were trained on the OpenSubtitles corpus using the fastText implementation of the skipgram algorithm. Performance comparable with (and in some cases exceeding) embeddings trained on non-conversational (Wikipedia) text is reported on standard benchmark evaluation datasets. A novel evaluation method of particular relevance to psycholinguists is also introduced: prediction of experimental lexical norms in multiple languages. The models, as well as code for reproducing the models and all analyses reported in this paper (implemented as a user-friendly Python package), are freely available at: https://github.com/jvparidon/subs2vec.

Transcript of Speech by Aslak Myhre, Director of National Library of Norway, at Fantastic Futures 2019: 2nd International Conference on AI for Libraries, Archives, and Museums

Transcript of Speech by Aslak Myhre, Director of National Library of Norway, at Fantastic Futures 2019: 2nd International Conference on AI for Libraries, Archives, and Museums

Retraction to: M111. QUANTITATIVE ASSESSMENT OF MANIA AND PSYCHOSIS DURING HOSPITALIZATION USING AUTOMATED ANALYSIS OF FACE, VOICE, AND LANGUAGE

Abstract Background A major challenge for reliable and effective mental health care is the lack of objective markers of illness. Computational approaches to measuring naturalistic behavior in clinical settings could therefore provide an objective backstop for mental health assessment and disease monitoring. This study aimed to train machine-learning (ML) classifiers to estimate conventional clinical measures of severe mental illness using quantitative metrics derived from computational analysis of facial and vocal behaviors. Methods Individuals hospitalized for any active psychotic condition were recruited to participate in up to ten recorded study visits, comprised of three segments. Each visit was captured using two synchronized HD webcams and cardioid microphones, to obtain high quality audiovisual (AV) data from both patient and interviewer. We performed automated facial action coding, vocal analysis, and speech transcription using publicly available software (e.g., openFace, openSmile, TranscribeMe). Results A total of 34 participants, participated in 66 sessions between 2015 and 2018, resulting in over 40 hours of AV recordings. In our visual and vocal analysis, we found that several features derived from face, voice, and use of language (i.e. eyebrow furrowing, eye widening, smile variability, characteristics of vowels) were both robustly measured using our approach, and allowed us to accurately estimate multiple symptom domains (i.e. mania, depression, psychosis) with (R= >0.7, p = <0.05). In our linguistic analysis, we found that abundance of power words (i.e. superiority, important) and lack of contextual language (i.e. yesterday, nearby) are highly indicative of positive psychotic symptoms with (R= +0.417, p = 0.002) and (R= -0.302, p = 0.028) respectively. Discussion Automated analysis of face, voice, and speech provides a number of robust behavioral markers sensitive enough to detect changes in psychopathology within individuals over time. Therefore, naturalistic, quantitative assessments can yield objective markers of mood and cognition that can be used to optimize both access and quality of treatments for a wide range of psychiatric conditions.

The Recognition of Persian Phonemes Using PPNet.

Background:In this paper, a novel approach is proposed for the recognition of Persian phonemes in the Persian consonant-vowel combination (PCVC) speech dataset. Nowadays, deep neural networks (NNs) play a crucial role in classification tasks. However, the best results in speech recognition are not yet as perfect as human recognition rate. Deep learning techniques show outstanding performance over many other classification tasks, such as image classification and document classification. Furthermore, the performance is sometimes better than a human. The reason why automatic speech recognition systems are not as qualified as the human speech recognition system, mostly depends on features of data which are fed to deep NNs.Methods:In this research, first, the sound samples are cut for the exact extraction of phoneme sounds in 50 ms samples. Then, phonemes are divided into 30 groups, containing 23 consonants, 6 vowels, and a silence phoneme.Results:The short-time Fourier transform is conducted on them, and the results are given to PPNet (a new deep convolutional NN architecture) classifier and a total average of 75.87% accuracy is reached which is the best result ever compared to other algorithms on separated Persian phonemes (like in PCVC speech dataset).Conclusion:This method not only can be used for recognizing mono-phonemes but it can also be adopted as an input to the selection of the best words in speech transcription