This thesis presents a study of several algorithms for enhancing speech degraded by additive noise in mobile cellular communications. The primary goal of this multi-stage study was to examine adaptive noise cancellation techniques in which one microphone is used to measure the speech plus noise signal and another microphone is used to form an estimate of the interfering background noise. The first stage of this research project involved the design and operation of a measurement system to gather dual channel audio samples in mobile radio environments for use in testing adaptive noise cancellation algorithms developed at Northeastern University. In the second phase of this research, several adaptive algorithms were used to implement noise cancellation systems which were applied to the measured speech signals. In the third phase of this research, several of the adaptive noise cancellation algorithms are compared and additional speech enhancement techniques are investigated. / Master of Science

This study evaluates a method for determining the quality of synthetic speech systems. The method involves the use of an auditory lexical decision task to assess the quality of synthetic speech generators relative to each other and to natural speech by using reaction time differences and error rates. Seven voices were evaluated; four synthesizers provided six voices (DECtalk 1.8 Perfect Paul, DECtaik 1.8 Beautiful Betty, DECtaik 2.0 Perfect Paul, DEC talk 2.0 Beautiful Betty, Votrax Personal Speech, Votrax Type'n'Talk) and natural speech provided the seventh voice. Both reaction times and error rates were higher for the low quality synthetic speech systems. The results document that the DECtalk can currently be considered a high quality synthesizer and that the Personal Speech and the Type'n'Talk can be considered low quality synthesizers. The results obtained by using this method can be explained by use of the Activation-Verification model (Paap, McDonald, Schvaneveldt, and Noel, 1986). Within the framework of this model, the results of this study suggest that the verification phase is the bottle-neck in processing words produced by synthetic speech generators. This interpretation suggests that by emphasizing the differences between different phonemes, to make them more uniquely identifiable, rather than concentrating on making them more "natural" might lead to improved results with synthesized speech. / Master of Science

Recent improvements in speech technology have made synthetic speech a viable I/O alternative. However, little research has focused on optimizing the various speech parameters which influence system performance. This study examined the effects of speech rate, message repetition, and the placement of information in a message. Briefly, subjects heard messages generated by a speech synthesizer and were asked to transcribe what they had heard. After entering each transcription, subjects rated the perceived difficulty of the preceding message, and how confident they were of their response. The accuracy of their response, system response time, and response latency were recorded. Transcription accuracy was best for messages spoken at 150 or 180 wpm and for messages repeated either twice or three times. Words at the end of messages were transcribed more accurately than words at the beginning of messages. Response latencies were fastest at 180 wpm with 3 repetitions and rose as the number of repetitions decreased. System response times were shortest when a message was repeated only once. The subjective certainty and difficulty ratings indicated that subjects were aware of errors when incorrectly transcribing a message. These results suggest that a) message rates should lie below 210 wpm, b) a repeat feature should be included in speech interface designs, and c) important information should be contained at the end of messages. / Master of Science

published_or_final_version / Speech and Hearing Sciences / Master / Master of Science in Audiology

Use of time-compressed speech stimuli has been found to be clinically effective in differential diagnosis of lesions of the temporal lobe. However, notably absent from the literature is information concerning performance of adults on time-compressed closed-set speech discrimination tasks. The goal of this study mas to compare performance of 12 males and 12 females between age 50 and age 70 on a time-compressed closed-set speech discrimination test against the performance of 12 males and 12 females between age 10 and age 28 on the same task. The Word Intelligibility by Picture Identification test (WPI) was presented in both non-compressed and time-compressed conditions to all subjects. Previous research suggests that a difference in performance between age groups and between males and females in the older age group should be expected. Average results indicated negligible differences between age or gender groups under any of the conditions tested. Additionally, the test yielded perfect or near perfect scores for all subjects in the non-compressed condition. Lack of differentiation of results suggests that the Word Intelligibility by Picture Identification CUIPI) may be insensitive to the discrimination disorders expected in older adults, that the subjects included in the study were atypical of older adults in general and therefore such discrimination disorders did not exist in the sample, or that the subjects in the study uiere able to apply some type of compensatory strategies which resulted in the unexpected performance.

Submitted by Filipe dos Santos (fsantos@pucsp.br) on 2018-04-05T12:25:01Z No. of bitstreams: 1 Tatiana Medeiros Deperon.pdf: 3084468 bytes, checksum: 25c018058ea863a026fad05f4842617b (MD5) / Made available in DSpace on 2018-04-05T12:25:01Z (GMT). No. of bitstreams: 1 Tatiana Medeiros Deperon.pdf: 3084468 bytes, checksum: 25c018058ea863a026fad05f4842617b (MD5) Previous issue date: 2018-02-27 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - CAPES / Introduction: The child's auditory experiences are influenced by the audibility of speech sounds. These experiences are modified by hearing loss, thereby altering developmental opportunities. Audibility, which can be measured by SII, has been shown to be a necessary condition for language development, since it allows access to the linguistic input. Several factors influence the development of these children and will moderate the relationship between their hearing loss and the results that each child will achieve. Vocabulary development has been shown to be a good indicator of language development. On this perspective, considering that audibility for speech sounds may not be sufficient for the development of oral language, it is necessary to analyze how these factors can interfere in the course and in ways to develop processes that facilitate rehabilitation. Objective: To identify the factors related to the development of oral language when audibility was established for speech sounds in hearing impaired children using who are hearing aid (HA) users. Methods: Sixty-five children aged 6 to 17 years old, HA users and based in the state of São Paulo, were evaluated for their receptive vocabulary performance (PPVT-4) and the relation to the following aspects: socioeconomic status, maternal education level, speech-language therapy, etiology, risk factors for hearing loss, age at diagnosis and intervention, use of the device, hearing and language categories, speech perception and audibility. Results: Within the population analyzed, audibility did not determine vocabulary performance. The variables: socioeconomic level, etiology, number of risk factors for hearing impairment, device usage and speech perception interfered in vocabulary performance. It was also verified that the chance of belonging to the group which scored greater vocabulary performance gets 84% smaller upon the increase of 1 risk factor, but increases in 55% when using the device for one extra hour per day. By relating chronological age and age equivalent in vocabulary performance, it was noted that the difference between the two indexes increases as subjects move away from the normal curve. It was observed that audibility tends to be higher in the group with higher vocabulary performance, and decreases with distance. In the group with good audibility and poor performance, late diagnosis and intervention and inconsistent device use may have been responsible for below-expected vocabulary performance. Conclusions: Factors such as: early diagnosis and intervention, device use, audibility and speech therapy are essential for language development. However, none of them stand-alone is sufficient to make this development happen, and aspects such as early diagnosis and intervention, device use, audibility and speech therapy are also fundamental for the language development to occur / dos sons de fala. Essas experiências são modificadas pela perda de audição, alterando consequentemente as oportunidades de desenvolvimento. A audibilidade, que pode ser medida pelo SII, tem se mostrado como uma condição necessária para o desenvolvimento de linguagem, pois permite o acesso ao input linguístico. Diversas variáveis influenciam o desenvolvimento destas crianças e irão moderar a relação entre a perda de audição e os resultados que cada criança vai alcançar. O desenvolvimento de vocabulário tem se mostrado um bom indicador do desenvolvimento de linguagem. Nesta perspectiva, considerando que a audibilidade para sons de fala pode não ser suficiente para o desenvolvimento da linguagem oral, cabe analisar quais e como estas variáveis podem interferir no percurso e desenvolver processos facilitadores na reabilitação. Objetivo: identificar os fatores relacionados ao desenvolvimento de linguagem oral quando estabelecida a audibilidade para sons de fala em crianças com deficiência auditiva usuárias de aparelho de amplificação sonora individual (AASI). Método: foram avaliadas 65 crianças entre 6 e 17 anos, usuárias de AASI de um centro de saúde auditiva no estado de São Paulo, quanto ao desempenho de vocabulário receptivo (PPVT-4) e sua relação com seguintes aspectos: nível socioeconômico, escolaridade materna, terapia fonoaudiológica, etiologia, fatores de risco para deficiência auditiva, idade no diagnóstico e na intervenção, uso do dispositivo, categorias de audição e linguagem, percepção de fala e audibilidade. Resultados: para a população estudada, a audibilidade não determinou o desempenho do vocabulário. As variáveis: nível socioeconômico, etiologia, número de fatores de risco para deficiência auditiva, uso do dispositivo e percepção de fala interferiram no desempenho de vocabulário. Verificou-se ainda que a chance de pertencer ao grupo com melhor desempenho de vocabulário fica 84% menor com o aumento de um fator de risco para deficiência, e aumenta 55% com o aumento de 1 hora/dia no uso do dispositivo. Ao relacionar a idade cronológica e a idade equivalente no desempenho de vocabulário, notou-se que a diferença entre os dois índices aumenta à medida que os sujeitos se afastam da curva de normalidade. Observou-se que a audibilidade tende a ser maior no grupo com desempenho de vocabulário superior, e diminui com a distância. No grupo com boa audibilidade e desempenho insatisfatório, o diagnóstico e intervenção tardios e uso inconsistente do dispositivo podem ter sido responsáveis pelo desempenho de vocabulário abaixo do esperado. Conclusões: fatores como: diagnóstico e intervenção precoces, uso do dispositivo, audibilidade e terapia fonoaudiológica são essenciais para o desenvolvimento de audição e linguagem. Entretanto, nenhum deles isoladamente é suficiente para ocasionar este desenvolvimento, e aspectos como: diagnóstico e intervenção precoces, uso do dispositivo, audibilidade e terapia fonoaudiológica também são fundamentais para que o desenvolvimento de audição e linguagem ocorra

The research offers an elaboration of genealogy as an approach to knowledge from Friedrich Nietzsche's original work on method (1887) processed by Michel Foucault (1971). It provides an interdisciplinary version that integrates theoretical and analytical contributions from the philosophy of knowledge, from exact and social sciences to artistic research. This work also takes into account those who have most explicitly enhanced the potentialities of geneaalogy such as Gilles Deleuze (1962) and, more recently, Carlo Sini (2007) and Giorgio Agamben (2010), but also embraces the reflections of researchers, from the past or present, whom I dare associate with the genealogy approach such as Max Weber (1922), Bruno Latour (2013), Fritjof Capra (2014), Diego Velázquez (1656) and the newer holistic and immersive approaches in digital art (Roy Ascott 2007). The result will be a "grid of intelligibility", an instrument of knowledge of the emerging phenomena that can be used for mappings and interdisciplinary networks and that, as in the original version (by Nietzsche and Foucault), intends to overcome the epistemological limits and disciplinary segmentation inherited by modernity. Such limits and segmentation are transferred in a special way in modern universities. For this reason, universities are not only the privileged object of this genealogical analysis, but also the field where its application is posed not only as a research practice or pedagogical tool, but also as a self-reflexive method on an organizational level. The genealogy of universities, therefore, is not just a speculative analysis, but a strategic and experimental choice, rather unusual, despite the vast literature available from a well-known text of Immanuel Kant (1798). Universities embody an intersection node among cultural, economic, political and technological trends, since their inception in the Middle Ages. They are the institutional entities that delegate big apparatus paradigm shifts that influence the approaches to knowledge of the people who live in a given social context in vehicular, transversal and vertical ways. Universities have always been the legitimised place to disclose knowledge approaches socially recognised. Their historical centrality and legitimacy has been renewed for more than five hundred years . In the contemporary world, their role is being compromised by global processes , neoliberalism and digitization in particular. This research will investigate the latter genealogically by focusing on the manifestations of resistance namely audit university and its development until the automation phase. In addition to contemporary authors such as Michael Power (1994), Laura Maran (2009), and Giovanni Leghissa (2012) - the theoretical framework will refer to Ivan Illich (1971), Edgar Morin (1999), but also Marshall McLuhan (1964) and those researchers that are currently involved in the analysis of the impact of media on the education system (Ben Fry, 2007, I. and M. Toru S. Vijay Kumar, 2008). It is an attempt to genealogically answer the question -What will the current audit universities become? This research has gone up to a drift that is more than a narrative exercise. It has pushed up to a prophecy that is only partly a fiction experiment namely automation in academia, which is the main research hypothesis. In an apocalyptic scenario automated universities represent an audit university involution, a result of hybridization among economic, technological, cultural and organizational phenomena. It is only by addressing this hybridization process that we can develop an alternative narrative. By following Antonio Caronia's (2008) approach, this research will use science-fiction language as a distortion of reality that allows creating, in Foucault's words, a doomsday scenario (a case of 'fiction historique') or an alternative perspective avoiding ideological risks. Along with contemporaries Derrick de Kerckhove (1998), Roy Ascott (2007), Marcello Giacomantonio (2007), Valeria Pinto (2012), Federico Butera (2007), I will summarize the works of Franz Kafka, Philip K. Dick and James Ballard, but also recent contributions from artistic and pedagogical research. Keywords: Genealogy, Grid of Intelligibility, Knowledge, Power, Control, Subjectivation, EHEA, Audit, Automation.

Made available in DSpace on 2016-04-28T18:22:44Z (GMT). No. of bitstreams: 1 Marcelo Henrique Barbosa.pdf: 2775707 bytes, checksum: 249f84c931dc60a5352e52b2546d37b4 (MD5) Previous issue date: 2013-10-15 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / This dissertation aims at discussing the relation between pronunciation and intelligibility in context of English as foreign language. It involves an investigation of the speech of Brazilian speakers of English, who learned the language during adulthood, through phonetic-acoustic and perception analyses. 24 native English speakers from 4 different countries performed the perceptive analyses: Australia, Canada, United Stated of America and United Kingdom. For the perceptive analyses performance 4 semantic descriptors have been used: Hard/Easy Understanding; Sounds Natural/Unnatural; Accurate/Inaccurate Intonation; Strong/Subtle Foreign Accent. For the development of this research 7 Brazilian and 4 North American subjects were recruited. The corpus consists of a monologue in English composed of 10 enunciations forming a phonetic inventory, which embody a variety of phonemes that normally represent pronunciation struggles for Brazilian speakers. The monologue was read and recorded at the Radio and TV studio-laboratory of PUCSP. Phonetic- acoustic analyses were done though a software called PRAAT and analyzed the enunciations that indicated more intelligibility s struggle through perceptive perspective. The results obtained points out that the majority of Brazilian subjects pronunciation interfered in the intelligibility of the speech productions; phonetic training contributed to approximate some Brazilians pronunciation to the North American subjects; some characteristics that should be considered within the learning-teaching process of the English language / Esta dissertação tem como objetivo discutir a relação entre pronúncia e inteligibilidade em contexto de inglês como língua estrangeira. Por meio de análises fonético-acústicas e perceptivas foi investigada a fala de sujeitos brasileiros falantes de língua inglesa que iniciaram sua aprendizagem desse idioma na fase adulta. A análise perceptiva foi efetuada por 24 falantes nativos de língua inglesa de 4 países diferentes: Austrália, Canadá, Estados Unidos da América e Reino Unido. Para a análise perceptiva utilizou-se quatro descritores de diferencial semântico a saber: Hard/Easy Understanding; Sounds Natural/Unnatural; Accurate/Inaccurate Intonation; Strong/Subtle Foreign Accent. O trabalho envolveu 7 sujeitos brasileiros e 4 estadunidenses. O corpus da pesquisa compreende um monólogo em inglês com 10 enunciados e constitui um inventário fonético que engloba uma variedade de fonemas que normalmente representam dificuldades de pronúncia para sujeitos brasileiros. O monólogo foi lido e gravado nas dependências do Estudio de Rádio e TV da PUC-SP. Análise fonético-acústica foi feita por meio do software PRAAT e analisou os enunciados que apresentaram maior dificuldade de inteligibilidade por parte da análise perceptiva. Os resultados obtidos apontam que a pronúncia da maioria dos sujeitos brasileiros interferiu na inteligibilidade das produções; o treinamento fonético contribuiu para a aproximação da pronúncia de um sujeito brasileiro aos sujeitos estadunidenses; algumas características de pronúncia que mereçam ser consideradas no ensino-aprendizagem de língua inglesa

No description available.

Unlike some Asian languages (e.g., Korean), English has lexical stress manifested by four acoustic features: duration, intensity, F0 (pitch), and vowel quality. Lexical stress has been known to have significant influences on native English speakers’ recognition of spoken words. According to Cutler (2015), lexical stress has both suprasegmental and segmental features: Suprasegmental features include duration, intensity, and F0 while vowel quality is considered a segmental feature. However, it is still unclear which lexical features are more responsible for spoken word recognition. This study examined which features, suprasegmental features or vowel quality of English, are a more significant influencer in spoken word recognition using English loanwords in Korean, which lack the prominence of any syllable realized by these features. Additionally, this study investigated the claimed advantage of the strong-weak stress pattern over a weak-strong pattern. To that end, two experiments were conducted. First, a parallel acoustic comparison was made between disyllabic English words and their corresponding English loanwords in Korean in order to investigate whether Korean has lexical stress features similar to those of English. 10 Korean and 10 English native speakers read 20 disyllabic words: the English loanwords in Korean by Korean participants and the source English words by American participants. The results showed that the differences of acoustic values between the syllables of the English words were significantly larger than those of the English loanwords. That is, the relative prominence of the stressed syllable over the unstressed syllable in English was not found in Korean. Additionally, the results indicated that Korean does not have a reduced vowel such as /ə/ in English, which is a critical feature of English vowel quality. In Experiment II, 16 English loanwords were used to create three versions of a spoken word recognition experiment, which was administered using the online survey platform, Qualtrics. Each version had a different type of manipulation: unmanipulated English loanwords, English loanwords with suprasegemental manipulation or English loanwords with vowel quality manipulation. 117 American English hearers identified the spoken words of one of the versions assigned to them; their success rates and reaction times (RT) were recorded. A binominal regression test was used for the analysis of success rates, and the Kruskal-Wallis H test for the response times. The results indicated that as far as success rates are concerned, both suprasegmental features and vowel quality play a role in recognizing spoken English words. However, when these two features were compared, vowel quality seemed to be a much stronger player. As for stress patterns, no significant differences were found in success rates across the three sets of manipulation. Moreover, this study did not find any significant difference in RTs either across the three manipulation sets or the two stress patterns. This study offered many applied implications in ESL, especially for teaching English pronunciation in Korea.