Show simple item record

dc.contributor.authorVan Niekerk, Daniel R.
dc.date.accessioned2018-03-02T11:52:37Z
dc.date.available2018-03-02T11:52:37Z
dc.date.issued2016
dc.identifier.citationDaniel van Niekerk, “Syllabification for Afrikaans speech synthesis”, in Proc. Pattern Recognition Association of South Africa and Robotics and Mechatronics International Conference (PRASA-RobMech), pp 31-36, Stellenbosch, South Africa, 2016. [http://engineering.nwu.ac.za/multilingual-speech-technologies-must/publications]en_US
dc.identifier.urihttp://ieeexplore.ieee.org/document/7813143/
dc.identifier.urihttp://www.demitasse.co.za/~demitasse/pubs/slides_prasa2016_afrttsdict.pdf
dc.identifier.urihttp://hdl.handle.net/10394/26486
dc.description.abstractThis paper describes the continuing development of a pronunciation resource for speech synthesis of Afrikaans by augmenting an existing pronunciation dictionary to include syllable boundaries and stress. Furthermore, different approaches for grapheme to phoneme conversion and syllabification derived from the dictionary are evaluated. Cross-validation experiments suggest that joint sequence models are effective at directly modelling pronunciations including syllable boundaries. Finally, some informal observations and demonstrations are presented regarding the integration of this work into a typical text-to-speech system.en_US
dc.description.sponsorshipThis work was partially funded by the Department of Arts and Culture of the Government of South Africa.en_US
dc.language.isoenen_US
dc.publisherPattern Recognition Association of South Africa and Mechatronics International Conferenceen_US
dc.subjectPronunciation dictionaryen_US
dc.subjectSyllabificationen_US
dc.subjectAfrikaansen_US
dc.subjectSpeech synthesisen_US
dc.subjectText-to-speechen_US
dc.titleSyllabification for Afrikaans speech synthesisen_US
dc.typePresentationen_US
dc.contributor.researchID21022658 - Van Niekerk, Daniël Rudolph


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record