Call for Paper - August 2022 Edition
IJCA solicits original research papers for the August 2022 Edition. Last date of manuscript submission is July 20, 2022. Read More

Comparative Study on Preprocessing Techniques on Automatic Speech Recognition for Tamil Language

Print
PDF
IJCA Proceedings on National Conference on Research Issues in Image Analysis and Mining Intelligence
© 2015 by IJCA Journal
NCRIIAMI 2015 - Number 2
Year of Publication: 2015
Authors:
S. Pannirselvam
G. Balakrishnan

S.pannirselvam and G.balakrishnan. Article: Comparative Study on Preprocessing Techniques on Automatic Speech Recognition for Tamil Language. IJCA Proceedings on National Conference on Research Issues in Image Analysis and Mining Intelligence NCRIIAMI 2015(2):25-28, June 2015. Full text available. BibTeX

@article{key:article,
	author = {S.pannirselvam and G.balakrishnan},
	title = {Article: Comparative Study on Preprocessing Techniques on Automatic Speech Recognition for Tamil Language},
	journal = {IJCA Proceedings on National Conference on Research Issues in Image Analysis and Mining Intelligence},
	year = {2015},
	volume = {NCRIIAMI 2015},
	number = {2},
	pages = {25-28},
	month = {June},
	note = {Full text available}
}

Abstract

Automatic Speech Recognition (ASR) is a flourishing and swift area for the conversion of acoustic signals acquired from human speech into various other forms such as text, actions, etc. , Conversion of Speech To Text (STT) is an incredible and challenging Task. In this paper, we present the study on comparing various digital representations for recording the speech, various pre-emphasis methods for removing the unwanted background noises from the recorded acoustics using suitable filtering techniques. The Filters also help to identify the formant waves for the betterment of syllable and phonetic identification in the subsequent operations for the detection of corresponding alphabetical text on STT Process. This study focuses only on the human speech source as in Tamil which is one among the various Dravidian Languages in India. The connection between oral and written form in Tamil is that individual phonetic segment of the speech denotes individual Tamil alphabets. This feature makes the recognition process as easier and accurate. The detection of location of each phoneme in the speech samples are based on accurate preprocessing outputs of the given speech signal. The last section of this paper shows the experimental results that compare the performance of some of the powerful pre-emphasis methods which are suitable for the Tamil utterance. Finally, we give the suggestions to prefer to use a particular method for the good segmentation.

References

  • Ronald. W. Schafer, Senior Member, IEEE, & Lawrence. R. Wbiner, Member, IEEE, "Digital Representations of Speech Signals", PROCEE-DINGS OF THE IEEE, vol. 63, no. 4, 1975.
  • G. Lakshmi Sarada, A Lakshmi, Hema A Murthy and T Nagarajan, "Automatic transcription of continuous speech into syllable-like units for Indian languages",Sadhana Vol. 34, Part 2, pp. 221–233,2009.
  • Abhishek Nandy," Pitch Detection of Speech Synthesis by Using Matlab",IOSR-JECE Volume 8, Issue 1, 2013.
  • Josh Reiss and Mark Sandler "Digital audio effects applied directly on a dsd bitstream", Proc. of the 7th Int. Conference on Digital Audio Effects, 2004.
  • Sarada, G. L. , Nagarajan, T. , Hema A. Murthy. , "MultipleFrame Size And Multiple Frame Rate Feature Extraction For Speech Recognition", SPCOM-2004, 2004.
  • Milind Kansara and Prof. Neeta Chapatwala "Noise Reduction from the Speech Signal using Wavelet Packet Transform" IJECSE,Volume2, Number 2, 2013.
  • Md. Mijanur Rahman,& Md. Al-Amin Bhuiyan "Dynamic Thresholding On Speech Segmentation",IJRET, Volume: 02 Issue: 09, 2013.
  • A. Montazeri,M. H. Kahaei &J. Postan,"A New Stable Adaptive IIR Filter For Active Noise Control Systems",2011
  • S. Jothilakshmi,S. Sindhuja &V. Ramilingam "Dravidian–Tamil Tts For Interactive Voice Response System",IJIRD, Vol 2,Issue 4, 2013.
  • Hanitha Gnanathesigar,"Tamil Speech Recognition using Semi Continuous Models",IJSRP,Vol 2, Issue 6, 2012.
  • Hanitha Gnanathesigar,"Tamil Speech Recognition using Semi Continuous Models",IJSRP,Vol 2 Issue 6 June 2012.