Zobrazeno 1 - 10
of 24
pro vyhledávání: '"Van Heerden, Charl"'
Publikováno v:
Speech Communication, 143, pp.10-20 (2022)
We propose a new framework to improve automatic speech recognition (ASR) systems in resource-scarce environments using a generative adversarial network (GAN) operating on acoustic input features. The GAN is used to enhance the features of mismatched
Externí odkaz:
http://arxiv.org/abs/2210.00721
Publikováno v:
Artificial Intelligence Research 2022
Mismatched data is a challenging problem for automatic speech recognition (ASR) systems. One of the most common techniques used to address mismatched data is multi-style training (MTR), a form of data augmentation that attempts to transform the train
Externí odkaz:
http://arxiv.org/abs/2202.07219
Autor:
Van Heerden, Charl Johannes
Higher-level features are considered to be a potential remedy against transmission line and cross-channel degradations, currently some of the biggest problems associated with speaker verification. Phoneme durations in particular are not altered by th
Externí odkaz:
http://hdl.handle.net/2263/25869
http://upetd.up.ac.za/thesis/available/etd-06262009-150945/
http://upetd.up.ac.za/thesis/available/etd-06262009-150945/
Efficient training of support vector machines and their hyperparameters / Charl Johannes van Heerden
Autor:
Van Heerden, Charl Johannes
As digital computers become increasingly powerful and ubiquitous, there is a growing need for pattern-recognition algorithms that can handle very large data sets. Support vector machines (SVMs), which are generally viewed as the most accurate classif
Externí odkaz:
http://hdl.handle.net/10394/11757
Autor:
Van Heerden, Charl Johannes.
Thesis (M.Eng.(Computer Engineering)--University of Pretoria, 2008.
Summaries in Afrikaans and English. Includes bibliographical references.
Summaries in Afrikaans and English. Includes bibliographical references.
Externí odkaz:
http://upetd.up.ac.za/thesis/available/etd-06262009-150945/
Publikováno v:
Language Resources and Evaluation, 2020 Jan 01. 54(1), 155-184.
Externí odkaz:
https://www.jstor.org/stable/48740862
Autor:
Kleynhans, Neil, Hartman, William, van Niekerk, Daniel, van Heerden, Charl, Schwartz, Rich, Tsakalidis, Stavros, Davel, Marelie
Publikováno v:
In Procedia Computer Science 2016 81:128-135
Publikováno v:
Language Resources and Evaluation, 2011 Jul 01. 45(3), 289-309.
Externí odkaz:
https://www.jstor.org/stable/41486044
We investigate the use of silhouette coefficients in cluster analysis for speaker diarisation, with the dual purpose of unsupervised fine-tuning during domain adaptation and determining the number of speakers in an audio file. Our main contribution i
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=od______1399::10a2d34a5bd01298262a0c40896fadbc
https://hdl.handle.net/10394/40042
https://hdl.handle.net/10394/40042
Autor:
Thirion, Jan Willem Frederick, Van Heerden, Charl Johannes, Giwa, Oluwapelumi, Davel, Marelie Hattingh
We present the design and development of a South African directory enquiries (DE) corpus. It contains audio and orthographic transcriptions of a wide range of South African names produced by first language speakers of four languages, namely Afrikaans
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=od______1399::94786f508a21882142bd8fcd2490c31e
https://hdl.handle.net/10394/36913
https://hdl.handle.net/10394/36913