SciELO - Scientific Electronic Library Online

 
vol.103 issue3-4 author indexsubject indexarticles search
Home Pagealphabetic serial listing  

Services on Demand

Article

Indicators

Related links

  • On index processCited by Google
  • On index processSimilars in Google

Share


South African Journal of Science

On-line version ISSN 1996-7489
Print version ISSN 0038-2353

S. Afr. j. sci. vol.103 n.3-4 Pretoria Mar./Apr. 2007

 

RESEARCH ARTICLES

 

Human and automatic accent identification of Nguni and Sotho Black South African English

 

 

Febe de WetI; Philippa LouwI; Thomas NieslerII

ICentre for Language and Speech Technology, University of Stellenbosch, Private Bag X1, Matieland 7602, South Africa
IIDepartment of Electrical and Electronic Engineering, University of Stellenbosch

 

 


ABSTRACT

It is well established that accent can have a detrimental effect on the performance of automatic speech recognition (ASR) systems. Whereas accents can be labelled in terms of a speaker's mother tongue, it remains to be determined if and when this distinction is appropriate for the development of ASR technology. This study compares the varieties of South African English produced by mother-tongue speakers of the Nguni and Sotho languages, who account for over 70% of the country's population. The aim of the investigation was to determine whether these two accent groups should be treated as a single variety by ASR systems, or whether it is better to consider them separately. To this end, two sets of experiments were carried out. First, a perceptual experiment was performed in which human listeners were required to classify different English accents. Subsequently, automatic speech recognition experiments were conducted to determine how the accuracy of an automatic accent identification system compares with these perceptual results, and whether the acoustic models benefit from the incorporation of Nguni/Sotho accent classifications. The results of the perceptual experiment indicated that most listeners could not correctly identify a speaker's mother tongue based on their English accent. This finding was supported by the results of the automatic accent identification and speech recognition experiments.


 

 

“Full text available only in PDF format”

 

 

REFERENCES

1. Goronzy S., Sahakyan M. and Wokurek W. (2001). Is non-native pronunciation modelling necessary? In Proceedings, Eurospeech 2001, pp. 309-312. Aalborg, Denmark.         [ Links ]

2. Aalburg S. and Hoege H. (2003). Approaches to foreign-accented speaker-independent speech recognition. In Proceedings, Eurospeech 2003, pp. 1489-1492. Geneva, Switzerland.         [ Links ]

3. Kessens J. (2006). Non-native pronunciation modelling in a command and control recognition task: a comparison between acoustic and lexical modelling. In Proceedings, First ISCA ITRW on Multilingual Speech and Language Processing (MULTILING), CDROM. Stellenbosch, South Africa.         [ Links ]

4. Coetzee-Van Rooy S. and van Rooy B. (2005). South African English: labels, comprehensibility and status. World Englishes 24(1), 1-19.         [ Links ]

5. de Klerk V. (2003). Towards a norm in South African Englishes: the case for Xhosa English. World Englishes 22(4), 463-481.         [ Links ]

6. van Rooy B. and van Huyssteen G. (2000). The vowels of BSAE: current knowledge and future prospects. S. Afr. J. Linguistics, Suppl. 38, 15-33.         [ Links ]

7. Wissing D. (2002). Black South African English: A new English? Some observations from a phonetic viewpoint. World Englishes 21(1), 129-144.         [ Links ]

8. van der Walt C. (2000). The international comprehensibility of varieties of South African English. World Englishes 19(2), 139-153.         [ Links ]

9. van Rooy B. (2004). Black South African English: phonology. In Handbook of Varieties of English, vol. 1, eds E. Schneider, K. Burridge, B. Kortmann, R. Mesthrie and C. Upton, pp. 943-952. Mouton, Berlin.         [ Links ]

10. Roux J.C., Louw P.H. and Niesler T.R. (2004). The African Speech Technology Project: an assessment. In Proceedings, 4th International Conference on Language Resources and Evaluation, Vol. 1, pp. 93-96. Lisbon, Portugal.         [ Links ]

11. Meierkord C. (2005). Black South African Englishes - towards a variationist account. Online: http//webdoc.sub.gwdg.de/edoc/ia/eese/artic25/meierk/1_2005.html (accessed 16/05/2007).         [ Links ]

12. Gough D. (1996). Black English in South Africa. In Focus on Africa, Varieties of English around the world, G15, pp. 53-77. John Benjamins, Amsterdam.         [ Links ]

13. Zissman M.A. and Berkling K.M. (2001). Automatic language identification. Speech Commun. 35, 115-124.         [ Links ]

14. Niesler T.R. and Willett D. (2006). Language identification and multilingual speech recognition using discriminatively trained acoustic models. In Proceedings, First ISCA ITRW on Multilingual Speech and Language Processing (MULTILING), CD-ROM. Stellenbosch, South Africa.         [ Links ]

15. Young S., Evermann G., Hain T., Kershaw D., Moore G. Odell J., Ollason D., Povey D., Valtchev V and Woodland P. (2002). The HTK Book, version 3.2.1. Cambridge University Engineering Department, Cambridge.         [ Links ]

16. Niesler T.R. and Louw P.H. (2004). Comparative phonetic analysis and phoneme recognition for Afrikaans, English and Xhosa using the African Speech Technology telephone speech databases. S. Afr. Computer J. 32, 3-12.         [ Links ]

17. Louw P.H. and de Wet F. (in press). The perception and identification of accent in spoken Black South African English. Southern African Linguistics and Applied Language Studies 25(1).         [ Links ]

 

 

Received 22 September 2006
Accepted 16 April 2007.

 

 

This article is a revised and extended version of a paper presented at the ISCA Tutorial and Research Workshop on Multilingual Speech and Language Processing (Stellenbosch, South Africa, 2006). Our work was supported by the National Research Foundation under grants FA2005022300010 and GUN2072874.
* Author for correspondence. E-mail: fdw@sun.ac.za

Creative Commons License All the contents of this journal, except where otherwise noted, is licensed under a Creative Commons Attribution License