A Study on Phonemic Analysis for the Recognition of Korean Speech

  • Authors:

  • Jeong Young Song;Min Wook Kil;Il Seok Ko

  • Affiliations:

  • Department of Computer Eng., Paichai University, 14 Yeonja 1Gil, DaeJeon, South Korea;Dept. of Medical Inform., Mun Kyung College, HoGyeMyun, Mun Kyung, South Korea;Dept. of Computer and Multimedia., Dongguk University, 707 Seokjang-dong, Gyeongju, South Korea

  • Venue:
  • ICCS '07 Proceedings of the 7th international conference on Computational Science, Part IV: ICCS 2007
  • Year:
  • 2007

Quantified Score

Hi-index 0.00

Visualization

Abstract

This paper is a study related to phonemic analysis of Korean speech. For speech recognition, the best method is that speech should be recognized as phonemic unit. However, to segment speech into phonemic units is not well-applied because of various changes of phonation. Therefore, I in this study try to arrange a phonemic system of Korean speech and segment speech into phonemic units on the base of the phonemic system of Hunminjeongeum, the Korean script. I also look for the features of each phoneme by observing the changes of frequency domain, mel bandwidth and mel capstrum of phoneme and aim to form the phonemic segmentation system of Korean speech.