218 resultados para Transcriptions
Resumo:
Traditional speech enhancement methods optimise signal-level criteria such as signal-to-noise ratio, but these approaches are sub-optimal for noise-robust speech recognition. Likelihood-maximising (LIMA) frameworks are an alternative that optimise parameters of enhancement algorithms based on state sequences generated for utterances with known transcriptions. Previous reports of LIMA frameworks have shown significant promise for improving speech recognition accuracies under additive background noise for a range of speech enhancement techniques. In this paper we discuss the drawbacks of the LIMA approach when multiple layers of acoustic mismatch are present – namely background noise and speaker accent. Experimentation using LIMA-based Mel-filterbank noise subtraction on American and Australian English in-car speech databases supports this discussion, demonstrating that inferior speech recognition performance occurs when a second layer of mismatch is seen during evaluation.
Resumo:
Traditional speech enhancement methods optimise signal-level criteria such as signal-to-noise ratio, but such approaches are sub-optimal for noise-robust speech recognition. Likelihood-maximising (LIMA) frameworks on the other hand, optimise the parameters of speech enhancement algorithms based on state sequences generated by a speech recogniser for utterances of known transcriptions. Previous applications of LIMA frameworks have generated a set of global enhancement parameters for all model states without taking in account the distribution of model occurrence, making optimisation susceptible to favouring frequently occurring models, in particular silence. In this paper, we demonstrate the existence of highly disproportionate phonetic distributions on two corpora with distinct speech tasks, and propose to normalise the influence of each phone based on a priori occurrence probabilities. Likelihood analysis and speech recognition experiments verify this approach for improving ASR performance in noisy environments.
Resumo:
A salutogenic approach explored themes of strength and well-being in life stories of Burmese refugees (N = 18) in Australia. Previous refugee studies have tended to focus on negative responses to traumatic events (e.g. posttraumatic stress disorder, depression). To widen the scope of refugee related research the focus of the current study was informed by a salutogenic perspective, exploring sources of strength that may facilitate well-being. Semi-structured narrative interviews explored: the participant's life before fleeing Burma, the journey of exile, and post-migration in Australia. Eight women and 10 men (Mage = 39 years) were interviewed and transcriptions analysis of narratives was conducted using Interpretative Phenomenological Analysis (IPA), with major themes being explicated. Super-ordinate themes pertaining to strength during times of hardship were identified and explicated as: support from interpersonal relationships, the pivotal role of values, a sense of future and agency, and reliance on spiritual or religious beliefs. Results indicate the existence of sources of strength that may contribute to human responses in times of hardship. Recognition and reflection of strengths may be incorporated into therapeutic and resettlement approaches for people from refugee backgrounds.
Resumo:
Background Many Internet-based treatments for depression and for alcohol misuse have a positive impact, yet little is known about how these treatments work. Most research on web-based interventions involves efficacy trials which, while important, offer little explanation about how people perceive and use online programs. Objective This study aimed to undertake a qualitative exploration of participants' experience, perceived impact and use of an integrated web-based program for comorbid depression and alcohol misuse. Specifically, it explored users' perspectives on the intensity of their treatment and the level of support they received. Methods Interviewees were drawn from participants in a randomised controlled trial of the OnTrack web-based treatment for depression and alcohol misuse, which compared Brief Self-Guided, Comprehensive Self-Guided and Comprehensive Therapist-Assisted versions of the program. Twenty-nine people (9–11 from each condition) completed semi-structured telephone interviews asking about their impressions and experiences with the program. Interview transcriptions were subject to a 6-step thematic analysis, employing a conceptual matrix to identify thematic differences across groups. Results Positive experiences and outcomes were more pronounced among participants receiving the comprehensive treatments than the brief one, but other responses were relatively consistent across conditions. A major theme was a wish for more individualisation and human contact, even in participants receiving emailed assistance. Some confused follow-up research assessments with therapist support. There was little correspondence between the perceived impact of the program and the amount reportedly completed, and some participants said they used strategies offline or completed exercises mentally. Conclusions This study highlighted discrepancies between how web-based treatments are intended to be used and how people actually engage with them. A challenge for the next wave of these interventions is the provision of individualised responses and coaching that retains an emphasis on self-management and constrains cost.
Resumo:
Genealogy; childhood in Tuebingen as the youngest of 14 brothers; student life in Tuebingen; professional career; because of his being Jewish no possibility to enter career as public prosecutor; active membership in National Liberal Party and in Free Mason Lodge; World War I; closing of his law office in 1933. Contains transcriptions of numerous congratulations for his golden wedding and his 80th birthday in 1937.
Resumo:
Typescript with handwritten corrections of a memoir, written in the form of a diary
Resumo:
Medication information is a critical part of the information required to ensure residents' safety in the highly collaborative care context of RACFs. Studies report poor medication information as a barrier to improve medication management in RACFs. Research exploring medication work practices in aged care settings remains limited. This study aimed to identify contextual and work practice factors contributing to breakdowns in medication information exchange in RACFs in relation to the medication administration process. We employed non-participant observations and semi-structured interviews to explore information practices in three Australian RACFs. Findings identified inefficiencies due to lack of information timeliness, manual stock management, multiple data transcriptions, inadequate design of essential documents such as administration sheets and a reliance on manual auditing procedures. Technological solutions such as electronic medication administration records offer opportunities to overcome some of the identified problems. However these interventions need to be designed to align with the collaborative team based processes they intend to support.
Resumo:
Copy of article on the Jewish cemetery in Lipperode, including transcriptions of tombstones and family trees for Ostheimer family; correspondence
Resumo:
Abstract (Mig or mej, själ or sjel? Problems and solutions in the transcription of Swedish song texts): In this article I am pointing out and discussing problems and solutions concerning phonetic transcription of Swedish song texts. My material consists of 66 Swedish songs phonetically transcribed. The transcriptions were published by The Academy of Finnish Art Song in 2009. The first issue was which level of accuracy should be chosen. The transcriptions were created to be clear at a glance and suitable for the needs of interpretation of non Swedish speaking singers. The principle was to use as few signs and symbols as possible without sacrificing accuracy. Certain songs were provided with additional information whenever there was a chance of misinterpretation. The second issue was which geographic variety of the language should be visible in the transcription, Standard Swedish or Finland-Swedish? The songs in the volume are a selection of well-known works that are also of international interest. Most were composed by Jean Sibelius (1865–1957), a substantial number of whose songs were based on poems written by Finland’s national poet, Johan Ludvig Runeberg (1804–1877). Thus I chose to use the variety of Swedish language spoken in Finland, in order to reflect the cultural origin of the songs. This variety differs slightly from the variety spoken in Sweden both on prosodic and phonetic level. In singing, the note-text gives the interpretor enough information about prosody. The differences concern mostly the phonemes. A fully consequent transcript was, however, difficult to make, due to vocal requirement. So, for example, in an unstressed final syllable the vowel was often indicated as a central vowel, which in singing is given a more direct emphasis than in a literal pronunciation, even if this central vowel does not occur in spoken Finland-Swedish.
Resumo:
An in vitro transcription system from Candida utilis is described. The template used is a hybrid plasmid containing Saccharomyces cerevisiae CYC1 promoter linked to a synthetic 377-bp G-minus casette (1). In vitro transcriptions are carried out in the presence of RNase. T1. Under these conditions only the transcripts that are resistant to RNase T1 accumulate. Using this protocol, it has been shown that in the absence of cytosolic factors RNA polymerase II (pol II) from C. utilis initiated RNA synthesis randomly. But both C. utilis and S. cerevisiae cell-free extracts could direct pol II from C. utilis to initiate transcription accurately. Results also indicated that the general transcription factors are functionally interchangeable between S. cerevisiae and C. utilis
Resumo:
A significant cost in obtaining acoustic training data is the generation of accurate transcriptions. For some sources close-caption data is available. This allows the use of lightly-supervised training techniques. However, for some sources and languages close-caption is not available. In these cases unsupervised training techniques must be used. This paper examines the use of unsupervised techniques for discriminative training. In unsupervised training automatic transcriptions from a recognition system are used for training. As these transcriptions may be errorful data selection may be useful. Two forms of selection are described, one to remove non-target language shows, the other to remove segments with low confidence. Experiments were carried out on a Mandarin transcriptions task. Two types of test data were considered, Broadcast News (BN) and Broadcast Conversations (BC). Results show that the gains from unsupervised discriminative training are highly dependent on the accuracy of the automatic transcriptions. © 2007 IEEE.
Resumo:
Analisa as influências das oposições minoritárias no processo legislativo no caso específico da PEC nº 40, de 2003 durante a tramitação na Câmara dos Deputados. Analisa, ainda, a atuação da oposição por meio dos instrumentos regimentais da Câmara dos Deputados e seus impactos sobre as variáveis tempo e mérito que envolvem a tramitação da proposta. Verifica a atuação da oposição nas Comissões e no Plenário da Câmara e quantifica as interferências sobre o processamento da proposta quanto ao tempo de tramitação e ao mérito.
Resumo:
Dr. Charles M. Breder, a well known ichthyologist, kept meticulous field diaries throughout his career. This publication is a transcription of field notes recorded during the Bacon Andros Expeditions, and trips to Florida, Ohio and Illinois during the 1930s. Breder's work in Andros included exploration of a "blue hole", inland ecosystems, and collection of marine and terrestrial specimens. Anecdotes include descriptions of camping on the beach, the "filly-mingoes" (flamingos) of Andros Island, the Marine Studios of Jacksonville, FL, a trip to Havana, and the birth of seahorses. This publication is part of a series of transcriptions of Dr. Breder's diaries. (PDF contains 55 pages)
Resumo:
A proposta dessa pesquisa é apontar possibilidades de escrita da história do município de Queimados, na Baixada Fluminense, a partir das memórias das lideranças que defenderam sua emancipação política, durante a década de 1980. Essas memórias, coletadas por meio de narrativas orais, se constituem no documento chave para essa análise, dado o caráter recente da emancipação desse município. A partir do estudo da transcrição dessas narrativas, coloca-se em discussão o processo de construção das memórias das lideranças da emancipação. No decurso desse processo, evidenciam-se as estratégias adotadas no sentido de afirmar a legitimidade da narrativa e reivindicação do reconhecimento pelas ações realizadas no passado. Da mesma forma, identificam-se silêncios, negações e hiatos, como aspectos que perpassam os discursos dos oito líderes políticos da Associação dos Amigos para o Progresso de Queimados, a AAPQ, especialmente quando são convidados a abordar as questões político-partidárias subjacentes ao campo político local que se configurou no período. Nesse intrincado cenário, evidencia-se o esforço pela construção das memórias da emancipação de Queimados, mais de duas décadas decorridas desde a conquista de sua autonomia política. Tais constatações nos permitiu trilhar alguns caminhos que inserem essa pesquisa no âmbito da História Pública, possibilitando um retorno de nosso trabalho a essa comunidade, sem perder de vista a ancoragem nos referenciais teóricos de Michael Pollak, de acordo com nossa prática enquanto historiadores.
Resumo:
This paper presents some developments in query expansion and document representation of our spoken document retrieval system and shows how various retrieval techniques affect performance for different sets of transcriptions derived from a common speech source. Modifications of the document representation are used, which combine several techniques for query expansion, knowledge-based on one hand and statistics-based on the other. Taken together, these techniques can improve Average Precision by over 19% relative to a system similar to that which we presented at TREC-7. These new experiments have also confirmed that the degradation of Average Precision due to a word error rate (WER) of 25% is quite small (3.7% relative) and can be reduced to almost zero (0.2% relative). The overall improvement of the retrieval system can also be observed for seven different sets of transcriptions from different recognition engines with a WER ranging from 24.8% to 61.5%. We hope to repeat these experiments when larger document collections become available, in order to evaluate the scalability of these techniques.