Text-to-speech synthesis can be an empowering communication tool in the hands of the print-disabled or augmentative and alternative communication user. In an effort to improve the naturalness of synthesised speech – and thus enhance the communication experience – we apply the natural language processing tasks of part-of-speech tagging and chunking to the text in the synthesis process. We cover the South African languages of (South African) English, Afrikaans, isiXhosa, isiZulu and Sepedi. The part-of-speech tagging delivers positive results for most of the languages; however, the chunking does not give any improvement in its current form.
Reference:
Schlünz, G.I., Dlamini, N. & Kruger, R.P. 2016. Part-of-speech tagging and chunking in text-to-speech synthesis for South African languages. In: Interspeech 2016, San Francisco, United States of America, 9 - 12 September 2016
Schlunz, G. I., Dlamini, N., & Kruger, R. P. (2016). Part-of-speech tagging and chunking in text-to-speech synthesis for South African languages. Curran Associates, Inc. http://hdl.handle.net/10204/11044
Schlunz, Georg I, Nkosikhona Dlamini, and Rynhardt P Kruger. "Part-of-speech tagging and chunking in text-to-speech synthesis for South African languages." (2016): http://hdl.handle.net/10204/11044
Schlunz GI, Dlamini N, Kruger RP, Part-of-speech tagging and chunking in text-to-speech synthesis for South African languages; Curran Associates, Inc; 2016. http://hdl.handle.net/10204/11044 .
Presented in: Interspeech 2016, San Francisco, United States of America, 9 - 12 September 2016. 'For any article published in Interspeech proceedings, ISCA grants each author permission to use the article in that author's dissertation or in institutional repositories (paper and/or electronic versions), provided that the article is correctly referenced (including page numbers and/or paper number)'
ResearchSpace
