GENERAL ENQUIRIES: Tel: + 27 12 841 2911 | Email: callcentre@csir.co.za

Browsing by Author "Barnard, E"

Sort by: Order: Results:

  • Joubert, G; Barnard, E (PRASA, 2005-11)
    The authors describe a system, based on open-source tools, that was developed in order to study the effect of network degenerations in Voice-over-Internet-Protocol applications on speech-recognition accuracy. Sophisticated ...
  • Kleynhans, NT; Barnard, E (Springer Verlag, 2014-10)
    Automatic speech recognition (ASR) technology has matured over the past few decades and has made significant impacts in a variety of fields, from assistive technologies to commercial products. However, ASR system development ...
  • Davel, MH; Barnard, E (2004-10)
    Bootstrapping techniques have significant potential for the efficient generation of linguistic resources such as electronic pronunciation dictionaries. The authors describe a system and an approach to bootstrapping for the ...
  • Davel, MH; Barnard, E (2004-10)
    The authors focus on factors related to the underlying rule-extraction algorithms, and demonstrate variants of the Dynamically Expanding Context algorithm, which are beneficial for this application. They show that continuous ...
  • Davel, MH; Van Heerden, C; Kleynhans, N; Barnard, E (The International Speech Communication Association, 2011-08)
    Spoken recordings that have been transcribed for human reading (e.g. as captions for audiovisual material, or to provide alternative modes of access to recordings) are widely available in many languages. Such recordings ...
  • Davel, M; Barnard, E (2006-11)
    When developing a language resource, there is generally a trade-off between the amount of effort invested in the resource creation process and the quality of the resulting resource. We (authors) argue that, in the developing ...
  • De Waal, A; Barnard, E (PRASA 2008, 2008-11)
    Topic models are unsupervised techniques that extract likely topics from text corpora, by creating probabilistic word-topic and topic-document associations. Evaluation of topic models is a challenge because (a) topic models ...
  • Davel, M; Barnard, E (2006-04)
    Various automated techniques can be used to generalise from phonemic lexicons through the extraction of grapheme-to-phoneme rule sets. These techniques are particularly useful when developing pronunciation models for ...
  • Botha, GR; Barnard, E (18th Annual Symposium of the Pattern Recognition Association of South Africa (PRASA), 2007-11)
    The authors investigate the factors that determine the performance of text-based language identification, with a particular focus on the 11 official languages of South Africa, using n-gram statistics as features for ...
  • Joubert, LJ; Davel, MH; Barnard, E; Zimu, V (PRASA, 2004-11)
    The need for a bootstrapping approach to the morphological decomposition of words in agglutinative languages such as isiZulu is motivated, and the complexities of such an approach are described. The authors then introduce ...
  • Barnard, E; Zarbian, S (2010-05)
    The authors investigate the acoustic realization of tone in continuous utterances in Sepedi (a language in the Southern Bantu family). Human labelers marked each of the 271 syllables in a 15-sentence corpus produced by a ...
  • Govender, N; Barnard, E; Davel, MH (Interspeech 2005, 2005-09)
    Much theoretical work has been done on the tonal structure of languages in the Bantu family. However, most of these studies are not supported by physical measurements, or even a consistent model for mapping from linguistic ...
  • Louw, A; Davel, M; Barnard, E (4th International Conference of the African Language Association of Southern Africa (ALASA), 2005-07)
    A general-purpose isiZulu text-to-speech (TTS) system was developed, based on the “Multisyn” unit-selection approach supported by the Festival TTS toolkit. The development involved a number of challenges related to the ...
  • Sharma Grover, A; Plauche, M; Barnard, E; Kuun, C (2009-04)
    This paper presents the work in the design of a SDS for the provision of health information to caregivers of HIV positive children. The authors specifically address the frequently debated question of input modality in ...
  • Sharma Grover, A; Plauche, M; Barnard, E (IEEE, 2009-04)
    This paper presents our work in the design of a SDS for the provision of health information to caregivers of HIV positive children. We specifically address the frequently debated question of input modality in speech systems; ...
  • Van Niekerk, DR; Barnard, E (18th Annual Symposium of the Pattern Recognition Association of South Africa (PRASA), 2007-11)
    When doing research into or building systems involving spoken language, one invariably relies on relevantly annotated speech data for analysis and incorporation into such systems. The authors investigate methods and ...
  • Badenhorst, J; Davel, MH; Barnard, E (PRASA, 2012-11)
    We improve on a piece-wise linear model of the trajectories of Mel Frequency Cepstral Coefficients, which are commonly used as features in Automatic Speech Recognition. For this purpose, we have created a very clean ...
  • De Waal, A; Barnard, E (PRASA 2010, 2010-11)
    Topic models explain a collection of documents with a small set of distributions over terms. These distributions over terms define the topics. Topic models ignore the structure of documents and use a bag-of-words approach ...
  • Zerbian, S; Barnard, E (Interspeech 2008, 2008-09)
    Tone in Bantu languages is rarely studied experimentally. This paper reports a production study which reveals the intricate interaction of tonal context and morphological structure in surface tone realization in Sepedi, a ...
  • Cloete, Jacob PL; Barnard, E; Patel, Hina M (2004-08)
    The authors report on research aimed at gaining an improved understanding of how issues such as language, literacy and previous exposure to technology affect citizens' ability to access government services of different ...