Advertisement

On the Use of Spectral Feature Fusions for Enhanced Performance of Malaysian English Accents Classification

  • Mohd Ali YusnitaEmail author
  • Murugesa Pandiyan Paulraj
  • Sazali Yaacob
  • Abu Bakar Shahriman
  • Rihana Yusuf
  • Shahilah Nordin
Conference paper

Abstract

Accent problem is a current issue that degrades the intelligibility and performance of speech recognition (ASR) systems. Despite English accents have been extensively researched in the United States, Britain, Australia, China, India, and Singapore, the study of Malaysian English (MalE) is still at infancy. There is till date, very limited evidence to corroborate how ethnically diverse accents in MalE of its three main ethnics can be identified from their speech signals. Most studies about MalE tackles issues from the view point of attitudinal studies and making use of human perceptual analysis. Instead, this paper presents experimental methods by means of acoustical analysis and machine learning techniques. In order to enhance the performance of accent classifier to classify the Malay, Chinese, and Indian accents this paper proposes fusion techniques of popularly known mel-frequency cepstral coefficients (MFCC) and linear prediction coefficients (LPC) with formants termed here as spectral feature fusions (SFFs). In these SFFs feature extractors, the main spectral features are fused with five usable formants and the extracted features are used to model K-nearest neighbors and artificial neural networks (ANN). Using independent test samples technique, gender-dependent accent classifiers were evaluated. Experimental results showed that the proposed SFFs surpassed the baseline features by 7.8 and 3.9 % increment of the classification rates for the LPC-formants and MFCC-formants fusions, respectively. The highest accuracies yielded for the fusion of MFCC and formants were 96.4 and 92.5 % on the male and female datasets. Speaking of LPC-formants fusion, the results were also promising, i.e., 92.6 and 88.8 % on the male and female datasets, respectively.

Keywords

Accent classification Formants Linear prediction coefficients Mel-frequency cepstral coefficients Malaysian english 

References

  1. Arslan LM (1996). Foreign accent classification in American English. Ph.D, Duke UniversityGoogle Scholar
  2. Arslan LM, Hansen JHL (1996) Language accent classification in American english. Speech Commun 18(4):353–367CrossRefGoogle Scholar
  3. Arslan LM, Hansen JHL (1997) Frequency characteristics of foreign accented speech. Paper presented at the 1997 IEEE international conference on acoustics. Speech Signal Process, 21–24 April 1997Google Scholar
  4. Basem HAA, Tan TP (2011) Non-native accent pronunciation modeling in automatic speech recognition. Paper presented at the 2011 international conference on asian language processing (IALP), Penang, Malaysia, 15–17 Nov 2011Google Scholar
  5. Chew LW, Seng KP, Ang LM, Ramakonar V, Gnanasegaran A (2011) Audio-emotion recognition system using parallel classifiers and audio feature analyzer. Paper presented at the third international conference on computational intelligence. Model Simul, Langkawi, Malaysia, 20–22 Sept 2011Google Scholar
  6. Davis S, Mermelstein P (1980) Comparison of parametric representations for monosyllabic word recognition in continuously spoken sentences. IEEE Trans Acoust Speech Signal Process 28(4):357–366CrossRefGoogle Scholar
  7. Deshpande S, Chikkerur S, Govindaraju V (2005) Accent classification in speech. Paper presented at the fourth IEEE workshop on automatic identification advanced technologies, 17–18 Oct 2005Google Scholar
  8. Diederich J, Pedersen C (2008) Accent in speech samples: support vector machines for classification and rule extraction. In: Diederich J (ed) Rule extraction from support vector machines, vol 80. Springer, Berlin, pp 205–226CrossRefGoogle Scholar
  9. Dupont S, Ris C, Deroo O, Poitoux S (2005) Feature extraction and acoustic modeling: an approach for improved generalization across languages and accents. Paper presented at the 2005 IEEE workshop onautomatic speech recognition and understanding, San Juan, 27–27 Nov 2005Google Scholar
  10. Fahlman SE (1988) An empirical study of learning speed in back-propagation networks. Carnegie Mellon UniversityGoogle Scholar
  11. Fohr D, Illina I (2007) Text-independent foreign accent classification using statistical methods. Paper presented at the 2007 IEEE international conference on signal processing and communications, 24–27 Nov 2007Google Scholar
  12. Furui S (2001) Digital speech processing, synthesis, and recognition, vol 7. CRCGoogle Scholar
  13. Garg M (2003) Linear prediction algorithms. Bombay, India: Indian Institute of Technology (IIT), pp 1–15Google Scholar
  14. Ghesquiere PJ, Compernolle DV (2002). Flemish accent identification based on formant and duration features. Paper presented at the The 2002 IEEE international conference on acoustics. Speech Signal Process (ICASSP) Orlando, FL, United States, 13–17 May 2002Google Scholar
  15. Ghorshi S, Vaseghi S, Yan Q (2008) Cross-entropic comparison of formants of British, Australian and American english accents. Speech Commun 50(7):564–579CrossRefGoogle Scholar
  16. Gill SK (1993) Standards and pedagogical norms for teaching english in Malaysia. World Englishes 12(2):223–238CrossRefGoogle Scholar
  17. Hanani A, Russell M, Carey MJ (2011) Speech-based identification of social groups in a single accent of British English by humans and computers. Paper presented at the 2011 IEEE international conference on acoustics, Speech Signal ProcessGoogle Scholar
  18. Humphries JJ, Woodland PC, Pearce D (1996) Using accent-specific pronunciation modelling for robust speech recognition. Paper presented at the fourth international conference on spoken language Philadelphia, PA, USA, 3–6 Oct 1996Google Scholar
  19. Liu WK, Fung P (1999) Fast accent identification and accented speech recognition. Paper presented at the IEEE international conference on acoustics, Speech Signal Process, Phoenix, AZ, USA 15–19 March 1999Google Scholar
  20. Looney CG (1997) Pattern recognition using neural network: theory and algorithm for engineers and scientists. Oxford University Press, New YorkGoogle Scholar
  21. Lowenberg P (1992) The marking of ethnicity in Malaysian english literature: nativization and its functions. World Englishes 11(2/3):251–258CrossRefGoogle Scholar
  22. Makhoul J (1975) Linear prediction: a tutorial review. Proc IEEE 63(4):561–580CrossRefGoogle Scholar
  23. Nair-Venugopal S (2000) English, identity and the Malaysian workplace. World Englishes, vol 19. Blackwell Publishers Ltd, Oxford, pp 205–213Google Scholar
  24. Nguyen P, Tran D, Huang X, Sharma D (2010) Australian accent-based speaker classification. Paper presented at the third international conference on knowledge discovery and data mining, Phuket, Thailand, 9–10 Jan 2010Google Scholar
  25. Phoon HS (2010) The phonological development of Malaysian english speaking chinese children: a normative study. Doctor of Philosophy, University of Canterbury. Communication Disorders, Christchurch, New Zealand. http://hdl.handle.net/10092/4336
  26. Picone JW (1993) Signal modeling techniques in speech recognition. Proc IEEE 81(9):1215–1247. doi: 10.1109/5.237532 CrossRefGoogle Scholar
  27. Pillai S, Mohd Don Z, Knowles G, Tang J (2010) Malaysian English: an instrumental analysis of vowel contrasts. World Englishes 29(2):159–172. doi: 10.1111/j.1467-971X.2010.01636.x CrossRefGoogle Scholar
  28. Pitton JW, Kuansan W, Biing-Hwang J (1996) Time-frequency analysis and auditory modeling for automatic recognition of speech. Proc IEEE 84(9):1199–1215CrossRefGoogle Scholar
  29. Rabiee A, Setayeshi S (2010) Persian accents identification using an adaptive neural network. Paper presented at the second international workshop on education technology and computer science, Wuhan, Hubei, China, 6–7 March 2010Google Scholar
  30. Rabiner L, Juang BH (1993) Fundamentals of speech recognition, vol 103. Prentice Hall, Englewood Cliffs, New JerseyGoogle Scholar
  31. Sangwan A, Hansen JHL (2012) Automatic analysis of Mandarin accented English using phonological features. Speech Commun 54(1):40–54CrossRefGoogle Scholar
  32. Teixeira C, Trancoso I, Serralheiro A (1996) Accent identification. Paper presented at the fourth international conference on spoken language, Philadelphia, PA, 3–6 Oct 1996Google Scholar
  33. Vallabha GK, Tuller B (2002) Systematic errors in the formant analysis of steady-state vowels. Speech Commun 38(1–2):141–160CrossRefGoogle Scholar
  34. Vieru B, de Mareüil PB, Adda-Decker M (2011) Characterisation and identification of non-native French accents. Speech Commun 53(3):292–310CrossRefGoogle Scholar
  35. Yusnita MA, Paulraj MP, Yaacob S, Shahriman AB, Saidatul A (2011) Malaysian english accents identification using LPC and formant analysis. Paper presented at the 2011 IEEE international conference on control system, Computing and Engineering (ICCSCE), Penang, Malaysia, 25–27 Nov 2011Google Scholar
  36. Yusnita MA, Paulraj MP, Yaacob S, Fadzilah MN, Shahriman AB (2013a) Acoustic analysis of formants across genders and ethnical accents in Malaysian english using ANOVA. Procedia Engineering 64(2013):385–394. doi: 10.1016/j.proeng.2013.09.111 CrossRefGoogle Scholar
  37. Yusnita MA, Paulraj MP, Yaacob S, Yusuf R, Shahriman AB (2013b) Analysis of accent-sensitive words in multi-resolution mel-frequency cepstral coefficients for classification of accents in Malaysian english. Int J Automot Mech Eng 7(2013):1053–1073CrossRefGoogle Scholar

Copyright information

© Springer Science+Business Media Singapore 2016

Authors and Affiliations

  • Mohd Ali Yusnita
    • 1
    Email author
  • Murugesa Pandiyan Paulraj
    • 2
  • Sazali Yaacob
    • 3
  • Abu Bakar Shahriman
    • 2
  • Rihana Yusuf
    • 1
  • Shahilah Nordin
    • 1
  1. 1.Faculty of Electrical EngineeringUniversiti Teknologi MARAKepala BatasMalaysia
  2. 2.School of Mechatronic EngineeringUniversiti Malaysia PerlisArauMalaysia
  3. 3.Universiti Kuala Lumpur Malaysian Spanish InstituteKuala LumpurMalaysia

Personalised recommendations