This dissertation details the implementation of a real-time, speaker-independent telephone auto attendant from first principles on limited quality speech data. An auto attendant is a computerized agent that answers the phone and switches the caller through to the desired person's extension after conducting a limited dialogue to determine the wishes of the caller, through the use of speech recognition technology. The platform is a computer with a telephone interface card. The speech recognition engine uses whole word hidden Markov modelling, with limited vocabulary and constrained (finite state) grammar. The feature set used is based on Mel frequency spaced cepstral coefficients. The Viterbi search is used together with the level building algorithm to recognise speech within the utterances. Word-spotting techniques including a "garbage" model, are used. Various techniques compensating for noise and a varying channel transfer function are employed to improve the recognition rate. An Afrikaans conversational interface prompts the caller for information. Detailed experiments illustrate the dependence and sensitivity of the system on its parameters, and show the influence of several techniques aimed at improving the recognition rate. / Dissertation (MEng (Computer Engineering))--University of Pretoria, 2006. / Electrical, Electronic and Computer Engineering / unrestricted

by Moon-Tong Chan. / Thesis submitted in: December 1997. / Thesis (M.Phil.)--Chinese University of Hong Kong, 1998. / Includes bibliographical references (leaves 123-124). / Abstract also in Chinese. / Chapter Chapter 1 --- Introduction --- p.1 / Chapter 1.1 --- The Data Set --- p.8 / Chapter Chapter 2 --- The Box-Jenkins Time Series Models --- p.15 / Chapter 2.1 --- The White-noise Process --- p.16 / Chapter 2.2 --- Stationarity of Time Series --- p.17 / Chapter 2.3 --- Differencing --- p.19 / Chapter 2.4 --- Seasonal ARIMA Models - SARIMA Models --- p.20 / Chapter 2.5 --- Intervention Models --- p.22 / Chapter 2.6 --- The Three Phases of ARMA Procedure --- p.24 / Chapter Chapter 3 --- Seasonal ARMA Models with Several Mean Levels --- p.38 / Chapter 3.1 --- Review of Linear Models --- p.40 / Chapter 3.1.1 --- Method of Weighted Least Squares --- p.41 / Chapter 3.2 --- The Proposed Model --- p.41 / Chapter 3.2.1 --- The Weightings --- p.43 / Chapter 3.2.2 --- Selection of Submodels --- p.45 / Chapter 3.2.3 --- Estimation of Model (3.4) --- p.46 / Chapter 3.3 --- Model Adequacy Checking --- p.55 / Chapter 3.3.1 --- Checking of Independence of Residuals --- p.56 / Chapter 3.3.2 --- Checking of Normality of Residuals --- p.58 / Chapter 3.4 --- Forecasting --- p.62 / Chapter Chapter 4 --- Comparison --- p.77 / Chapter 4.1 --- Similarities and Differences Between the Two Models --- p.78 / Chapter 4.2 --- Model Comparative Criterion --- p.81 / Chapter 4.2.1 --- Model Fitting Comparison --- p.82 / Chapter 4.2.2 --- Model Forecasting Comparison --- p.83 / Chapter 4.3 --- Conclusion --- p.90 / Chapter 4.4 --- Generation of Predicted Hourly Calls --- p.91 / Chapter 4.5 --- Extension --- p.92 / Appendix A --- p.97 / Appendix B --- p.105 / Appendix C --- p.122 / References --- p.123

Community-based Question Answering (CQA) services enable members to ask questions and have them answered by the community. These services have the potential of rapidly creating large archives of questions and answers. However, their information is rarely exploited. This thesis presents a new statistical topic model for modeling Question-Answering archives. The model explicitly captures topic dependency and correlation between questions and answers, and models differences in their vocabulary. The proposed model is applied for the task of Question Answering and its performance is evaluated using a dataset extracted from the programming website Stack Overflow. Experimental results show that it achieves improved performance in retrieving the correct answer for a query question compared to the LDA model. The model has also been applied for Automatic Tagging and comparisons with LDA show that the new model achieves better clustering performance for larger numbers of topics.

Thesis (M.A.)--University of Hong Kong, 1987. / Also available in print.

by Chan Kit Ping, Wendy. / Includes questionnaire in Chinese. / Thesis (M.B.A.)--Chinese University of Hong Kong, 1994. / Includes bibliographical references (leaves 50-52). / ABSTRACT --- p.ii / TABLE OF CONTENTS --- p.iv / LIST OF FIGURES --- p.vi / LIST OF TABLES --- p.vii / ACKNOWLEDGEMENT --- p.viii / Chapter / Chapter I. --- INTRODUCTION --- p.1 / Chapter II. --- FACTORS THAT CONTRIBUTE TO THE DEVELOPMENT OF TELEPHONE BANKING SERVICE IN HONG KONG --- p.7 / High Telephone Usage in Hong Kong --- p.7 / Telephone Banking Service as Differentiation Weapon --- p.8 / Hectic Life Style of Hong Kong People --- p.8 / High Property Prices in Hong Kong --- p.8 / Labor Shortage in Hong Kong --- p.9 / Chapter III. --- INDUSTRY REVIEW --- p.10 / Development of Telephone Banking Service in Hong Kong --- p.10 / Procedure of Using Telephone Banking Service --- p.12 / Variants of Telephone Banking Services --- p.16 / Chapter IV. --- LITERATURE REVIEW --- p.20 / Adoption of a Service Innovation --- p.21 / Adoption of New Banking Technology --- p.21 / Hypotheses Setting --- p.25 / Chapter V. --- RESEARCH METHOD --- p.27 / Research Information Needed --- p.27 / Research Design --- p.29 / Demographic Characteristics of the Sample --- p.33 / Chapter VI. --- RESEARCH ANALYSIS --- p.34 / Awareness of Telephone Banking Service in Hong Kong --- p.34 / Way of Learning about Telephone Banking Service --- p.34 / Adoption of the Service --- p.35 / Reasons for Using the Service --- p.35 / Frequency of Using Telephone Banking Services --- p.36 / Most Frequently Used Services --- p.36 / Satisfaction Level of Users --- p.37 / Reasons for Not Using the Service --- p.37 / Attitude Towards Telephone Banking Service --- p.37 / Psychographic Characteristics of Users vs Non-users --- p.38 / Chapter VII. --- RECOMMENDATIONS --- p.40 / Ways to Recruit New Users --- p.40 / Ways to Encourage More Usage from Existing Users --- p.43 / Operational Recommendations --- p.44 / Chapter VIII. --- LIMITATIONS AND SUGGESTIONS --- p.46 / Questionnaire Setting --- p.45 / Sample Size --- p.47 / The Use of Personal Questions --- p.48 / Suggestions for Future Researches --- p.48 / BIBLIOGRAPHY --- p.50 / APPENDIX --- p.53 / Questionnaire --- p.53 / Figures 1-13 --- p.63-80 / Tables 1-13 --- p.81-93

Thesis (MPhil (General Linguistics))--University of Stellenbosch, 2009.

Made available in DSpace on 2016-04-27T18:12:26Z (GMT). No. of bitstreams: 1 Aline Tavares Pimentel.pdf: 623989 bytes, checksum: 0ca4614b775cdc0e01ab2e8e27d5f7e0 (MD5) Previous issue date: 2007-10-09 / Background: the professional voice, the verbal behavior and the mode as the communication interpersonal can be define the relations of job. To relate the verbal expressivity about stress, of form that if can think about who these aspects intervene is something recent in the speech-therapy. The telemarketing and the conditions of job have been explored in the research, the field of the professional voice. Objective: to investigate the relation between the symptoms of stress and the characteristics of verbal expressivity in active teleoperators of a central office of telemarketing of Salvador. Method: 52 teleoperators of the active sector had been selected, in the period between March and June of 2007. In this group, two protocols had been applied. First one investigated aspects about general health and the second the protocol approached aspects relative the occupational questions. Also the questionnaire Job Stress Scale for the verification was applied of stress in the work environment. Were select the records of the two teleoperators that had presented high consuming and low consuming for analyzes of the verbal expressivity for a speech therapy specialist in voice. Results: high consuming in 50% of the teleoperators was verified, haven t significance statistics between perception of the teleoperators and stress. The teleoperator with high consuming related great number of health problems. About your verbal expressivity this was positive, a time that presented intonations variations, adequate pauses and emphases and contextualized inflection. The teleoperator with low consuming, related little symptoms of health problems and about your verbal expressivity can be considered more negative, she show less emphases and phrases. Conclusion: The stress affected positively the verbal expressivity of the teleoperator with high consuming and refusal of the teleoperator with low consuming. Suggests work groups and discussion about stress inside the company, similar to increase the perception and assistant in the development of management for stress / Introdução: a voz profissional, a expressividade oral e o modo como a comunicação interpessoais pode definir as relações de trabalho. Relacionar a expressividade oral com estresse, de forma que se possa pensar em como esses aspectos interferem é algo recente na Fonoaudiologia. O teleatendimento e as condições de trabalho têm sido temas explorados nas pesquisas, no campo da voz profissional. Objetivo: investigar a relação entre os sintomas de estresse e as características de expressividade oral em teleoperadoras ativas de uma central de teleatendimento de Salvador. Método: foram selecionados 52 teleoperadoras do setor ativo, no período entre março e junho de 2007. Nesse grupo foram aplicados dois protocolos. O primeiro investigou aspectos da saúde geral e o segundo protocolo abordou aspectos relativos a questões ocupacionais. Também foi aplicado o questionário Job Stress Scale para a verificação do estresse no ambiente de trabalho. Foram selecionadas as gravações de duas teleoperadoras que apresentaram alto desgaste e baixo desgaste para analise da expressividade oral por uma fonoaudióloga especialista. Resultados: verificou-se alto desgaste em 50% das teleoperadoras, não teve significância estatística entre a percepção das teleoperadoras e estresse. A teleoperadora com alto desgaste referiu grande número de problemas de saúde. Quanto à expressividade oral esta foi positiva, uma vez que apresentou variações entonacionais, pausas e ênfases adequadas e inflexão contextualizada. A teleoperadora com baixo desgaste, referiu menos sintomas de problemas de saúde e quanto a expressividade oral pode ser considerada mais negativa, por apresentar menos ênfases e frases mais curtas. Conclusão: O estresse afetou positivamente a expressividade oral da teleoperadora com alto desgaste e negativa a da teleoperadora com baixo desgaste. Sugere-se grupos de trabalho e discussão sobre o estresse na empresa, a fim de aumentar a percepção e auxiliar no desenvolvimento de gerenciamento para o estresse

Introdução: A exposição ocupacional ao ruído tem sido um tema estudado para diferentes ocupações. Na atividade de teleatendimento também tem sido referido como um fator de risco com repercussões à saúde. Objetivo: Identificar a presença de perdas auditivas em operadores de teleatendimento da central de reservas de uma empresa aérea. Métodos: Realizado estudo epidemiológico transversal e longitudinal. O estudo transversal compreendeu os últimos audiogramas de 589 operadores distribuídos de acordo com o sexo, faixa etária (18 a 30, 31 a 43, mais de 44 anos), e tempo na função (01 a 35, 36 e 60, mais 61 meses). Foram avaliados os audiogramas de 387 operadores que estavam na empresa há mais de 3 anos, através de um estudo longitudinal, comparando os audiogramas referenciais e finais no período de 1999 a 2006. Resultados: Identificou-se o predomínio do sexo feminino, prevalência de perdas auditivas de 6,0% com 3,1% de PAIR e 2,9% por outras causas. Para as perdas auditivas sexo não se mostrou um fator de risco significativo (p>0,05). Observa-se um risco maior para as faixas etárias de 31 a 43 e de 41 a 55 anos de idade e tempo de serviço maior de 61 meses com chance de 3,35 (IC95%: 1,31 8,57) quando comparados com o tempo de serviço menor do que 36 meses. A comparação entre as médias mostrou um aumento significante dos limiares com exceção da freqüência de 1K na orelha direita. Conclusões: Operadores de teleatendimento na empresa estudada apresentaram rebaixamento dos limiares auditivos em todas as freqüências. A exposição ocupacional por ruído ambiental e uso de headset dos operadores de teleatendimento necessita de estudos longitudinais detalhados para definição das repercussões auditivas e extra-auditivas / Introduction: Occupational noise exposure has been a subject studied for various occupations. At call centers, it has also been referred to as a healthhazardous risk factor. Aim: Identify the presence of hearing loss in operators of an airline company call center. Methodology: Cross-section and longitudinal epistemological study has been carried out. The cross-section study considered the latest audiograms of 589 operators divided according to gender, age (18 to 30, 31 to 43, over 44), and employment time (01 to 35, 36 to 60, over 61 months). The longitudinal study considered the audiograms of 387 operators who have been with the company for over 3 years, comparing referencial and final audiograms of the period between 1999 and 2006. Results: A predominance of 6.0% hearing loss, with 3.1% of NIHL and 2.9% by other causes, has been noticed in females. For hearing losses, gender has not turned out to be a significant risk factor (p>0.05). There is a greater risk for ages between 31 and 43, and 41 and 55, having been with the company for over 61 months, of 3.35 (IC95%: 1.31 - 8.57) when compared to those who have been with the company for less than 36 months. The comparison between the averages has shown a significant rise of threshold with the exception of 1K frequency in the right ear. Conclusions: Call center operators in the airline company at stake have shown lowering of hearing threshold in all frequencies. Noise-induced occupational exposure and the use of headsets by call center operators demand further, more detailed longitudinal studies to define the hearing and extra-hearing repercussions

Introdução: A exposição ocupacional ao ruído tem sido um tema estudado para diferentes ocupações. Na atividade de teleatendimento também tem sido referido como um fator de risco com repercussões à saúde. Objetivo: Identificar a presença de perdas auditivas em operadores de teleatendimento da central de reservas de uma empresa aérea. Métodos: Realizado estudo epidemiológico transversal e longitudinal. O estudo transversal compreendeu os últimos audiogramas de 589 operadores distribuídos de acordo com o sexo, faixa etária (18 a 30, 31 a 43, mais de 44 anos), e tempo na função (01 a 35, 36 e 60, mais 61 meses). Foram avaliados os audiogramas de 387 operadores que estavam na empresa há mais de 3 anos, através de um estudo longitudinal, comparando os audiogramas referenciais e finais no período de 1999 a 2006. Resultados: Identificou-se o predomínio do sexo feminino, prevalência de perdas auditivas de 6,0% com 3,1% de PAIR e 2,9% por outras causas. Para as perdas auditivas sexo não se mostrou um fator de risco significativo (p>0,05). Observa-se um risco maior para as faixas etárias de 31 a 43 e de 41 a 55 anos de idade e tempo de serviço maior de 61 meses com chance de 3,35 (IC95%: 1,31 8,57) quando comparados com o tempo de serviço menor do que 36 meses. A comparação entre as médias mostrou um aumento significante dos limiares com exceção da freqüência de 1K na orelha direita. Conclusões: Operadores de teleatendimento na empresa estudada apresentaram rebaixamento dos limiares auditivos em todas as freqüências. A exposição ocupacional por ruído ambiental e uso de headset dos operadores de teleatendimento necessita de estudos longitudinais detalhados para definição das repercussões auditivas e extra-auditivas / Introduction: Occupational noise exposure has been a subject studied for various occupations. At call centers, it has also been referred to as a healthhazardous risk factor. Aim: Identify the presence of hearing loss in operators of an airline company call center. Methodology: Cross-section and longitudinal epistemological study has been carried out. The cross-section study considered the latest audiograms of 589 operators divided according to gender, age (18 to 30, 31 to 43, over 44), and employment time (01 to 35, 36 to 60, over 61 months). The longitudinal study considered the audiograms of 387 operators who have been with the company for over 3 years, comparing referencial and final audiograms of the period between 1999 and 2006. Results: A predominance of 6.0% hearing loss, with 3.1% of NIHL and 2.9% by other causes, has been noticed in females. For hearing losses, gender has not turned out to be a significant risk factor (p>0.05). There is a greater risk for ages between 31 and 43, and 41 and 55, having been with the company for over 61 months, of 3.35 (IC95%: 1.31 - 8.57) when compared to those who have been with the company for less than 36 months. The comparison between the averages has shown a significant rise of threshold with the exception of 1K frequency in the right ear. Conclusions: Call center operators in the airline company at stake have shown lowering of hearing threshold in all frequencies. Noise-induced occupational exposure and the use of headsets by call center operators demand further, more detailed longitudinal studies to define the hearing and extra-hearing repercussions

published_or_final_version / Language Studies / Master / Master of Arts