Total views : 180

Analysis of Vocal Tract Shape Variability based on Formant Frequency Ratio at Various Conditions of Vowels for Indian English Speakers

Affiliations

  • Electronics Engineering, Jain University, Bangalore – 560069, Karnataka, India
  • Akshaya Institute of Technology, Tumkur – 572106, Karnataka, India

Abstract


The paper presents the notability of variation of vocal tract shape and formant frequency with its adjacent ratios, for the Indian English speakers for the five vowels of the English language. we have estimated the vocal tract shape of the Indian English speakers. For the estimation we have incorporated Autoregressive Model and the same for the formant frequency. The speech samples are considered with three various utterances namely the consumption of Ice cold water, with time relaxation of five minutes compared with normal recordings of the vowels namely /a/, /e/, /I/, /o/ and /u/. These utterances are recorded for 20 individual iterations. the vocal tract shape estimation and formant frequency estimation are done on the Matlab platform. the vocal tract shape of first vowel /a/ comes to normal shape rapidly with time lapse of five minutes. the vocal tract shape of vowel /e/ shrinks after consumption of ice cold water and slowly attains the normal shape. The vowel /I/ and /or/ vocal tract shape changes linearly for all the conditions considered, whereas the vocal tract shape of /u/ compresses for ice cold water and retains to be same even after a time lapse of five minutes. All the above variations are done by considering the vocal tract length of 17cm and it is modelled according to lossless uniform tube model. these outcomes are used for observation of vocal tract infection among the speech disorder patients, it can be adopted for the user authentication system by considering it as a vocal tract signature.

Keywords

English Vowels, Formant Frequency, Indian English, Vocal Tract Shape, Vowels.

Full Text:

 |  (PDF views: 130)

References


  • Anil Kumar C, Shiva Prasad KM, Manjunatha MB, KodandaRamaiah GN. Basic Acoustic Features Analysis of Vowels And C-V-C of Indian English Language. ITSITEEE. 2015; 3(1): 20–3.
  • Shiva Prasad KM, Anil Kumar C, Manjunatha MB, KodandaRamaiah GN. Gender based Acoustic Features and Spectrogram analysis for kannada Phonetics. ITSITEEE. 2015; 3(1): 16–9.
  • Anil Kumar C, Manjunatha MB, Kodanda Ramaiah GN. Formant Frequency Based Analysis of English vowels for various Indian Speakers at different conditions using LPC and default AR modelling. International Journal of Computer Science and Information Security (IJCSIS). 2016; 14 Special Issue.
  • Powen Ru, Taishih Chi, and S Shamma. The synergy between speech production and perception. The Journal of Acoustical Society of America. 2003 Jan; J. Makhoul, “Linear Prediction: A Tutorial Review,” Proc. IEEE, Vol. 63,pp 561-580,1975.
  • Rexy RR, Rosini RR, Mani Sankar R. A Novel Speech Recognition System using Hidden Markov Model. Indian Journal of Science and Technology. 2015 Nov; 8(32).
  • Shiva Prasad KM, Kodanda Ramaiah GM, Manjunatha MB. Backend Tools for Speech Synthesis in Speech Processing. Indian Journal of Science and Technology. 2017 January; 10(1). Crossref
  • Anil Kumar C, Shiva Prasad KM, Manjunatha MB, KodandaRamaiah GN. Vocal Tract shape estimation of Vowels and C-V-V-C for diversified Indian English Speakers. IEEE International Conference on Electrical, Electronics, Signals, Communications and Optimization. 2015 Jan; 1–7. Crossref
  • Shiva Prasad KM, Anil Kumar C, KodandaRamaiah GN, Manjunatha MB. Speaker based vocal tract shape estimation for kannada vowels. IEEE International Conference on Electrical, Electronics, Signals, Communications and Optimization. 2015. 1–6. Crossref
  • Wankhede NS, Shah MS. Investigation on optimum parameters for LPC based vocal tract shape estimation. IEEE-Emerging Trends in Communication, Control, Signal Processing and Computing Applications C2SPCA. 2013; 1–6.
  • Sohal B, Sandeep K. A HMM Integrated SVM Model for Hindi Speech Recognition. Indian Journal of Science and Technology. 2016 December; 9(47). Crossref 11. Scordillis M, Gowdy JN. Effects of the vocal tract shape on the spectral tilt of the glottal pulse wave form. IEEE Transactions. 1990; 86–9.
  • Sajeer K, Rodrigues P. Novel Approach of Implementing Speech Recognition using Neural Networks for Information Retrieval. Indian Journal of Science and Technology. 2015 December; 8(33).
  • khodai-joopaari M, Clermont F, Barlow M. Speaker variability on a continuum of spectral sub-bands from 297-speakers’non- contemporaneous cepstra of japans vowels. Proceedings of the 10th Australian international conference on speech scicence and technology. 2004.
  • Dong-Ill Kim, Byung-Cheol Kim. Speech Recognition using Hidden Markov Models in Embedded Platform. Indian Journal of Science and Technology. 2015 December; 8(34). Crossref December 2015.
  • Shiva Prasad KM, Anil Kumar C, Manjunatha MB, KodandaRamaiah GN. Various front end tools for digital speech. Processing in IEEE. 2015; 905–11. Print ISBN: 9789-3805-4415-1.
  • Shiva Prasad KM, Kodanda Ramaiah GN, Manjunatha MB. Speech Features Extraction Techniques for Robust Emotional Speech Analysis/Recognition. Indian Journal of Science and Technology, 2017 January; 10(3): . Crossref
  • Kodandaramaiah GN, Giriprasad MN, Mukundarao M. Implementation of LPC based vocal tract shape estimation for vowels. The Technology world quarterly journal. 2010 march-April; 90(2): 97–102. ISSN 2180-1614.
  • Vallabha G, Tuller B. Choice Of Filter Order In Lpc Analysis Of Vowels. Sound to Sense. 2004 June; at MIT.
  • Shah MS, Pandey PC. Estimation of vocal tract shape for VCV syllables for a speech training aid. Proceedings of 27th International Conference on IEEE Engineering Medicine and Biology Society. 2005; 6642–5. Crossref
  • Raut SV, Panthangi LC, Akhil BG, Syed Faisal Ali, Sanjay HS and Bhargavi S. Classifcation of Sex based Speech Differentiation in Healthy Human Beings based on Voiced and Unvoiced Components. Indian Journal of Science and Technology. 2017 January; 10(1). Crossref

Refbacks

  • There are currently no refbacks.


Creative Commons License
This work is licensed under a Creative Commons Attribution 3.0 License.