DOI Number : 10.5614/itbj.ict.res.appl.2014.8.2.6
Hits : 22

New Grapheme Generation Rules for Two-Stage Modelbased Grapheme-to-Phoneme Conversion

Seng Kheang1, Kouichi Katsurada1, Yurie Iribe2 & Tsuneo Nitta1

1Toyohashi University of Technology, 1-1 Tempaku, Toyohashi, Aichi 441-8580, Japan ,
2Aichi Prefectural University, 1522-3 Ibaragabasama, Nagakute, Aichi 480-1198, Japan

Abstract. The precise conversion of arbitrary text into its corresponding phoneme sequence (grapheme-to-phoneme or G2P conversion) is implemented in speech synthesis and recognition, pronunciation learning software, spoken term detection and spoken document retrieval systems. Because the quality of this module plays an important role in the performance of such systems and many problems regarding G2P conversion have been reported, we propose a novel two-stage model-based approach, which is implemented using an existing weighted finite-state transducer-based G2P conversion framework, to improve the performance of the G2P conversion model. The first-stage model is built for automatic conversion of words to phonemes, while the second-stage model utilizes the input graphemes and output phonemes obtained from the first stage to determine the best final output phoneme sequence. Additionally, we designed new grapheme generation rules, which enable extra detail for the vowel and consonant graphemes appearing within a word. When compared with previous approaches, the evaluation results indicate that our approach using rules focusing on the vowel graphemes slightly improved the accuracy of the out-of-vocabulary dataset and consistently increased the accuracy of the in-vocabulary dataset.

Keywords: grapheme generation rules (GGR); combined grapheme-phoneme information; two-stage model; grapheme-to-phoneme (G2P); automatic text-to-phonetic transcription.

Download Article
Bahasa Indonesia | English


Begin on 10 October 2014 this website is no longer activated for article process in Journal of Mathematical and Fundamental Sciences, Journal of Engineering and Technological Sciences, Journal of ICT Research and Applications and Journal of Visual Art and Design. The next process will be proceeded under new website at

For detail information please contact us to:

       ITB Journal Visitor Number #9628379       
       Jl. Tamansari 64, Bandung 40116, Indonesia Visitor IP Address #       
       Tel : +62-22-250 1759 ext. 121 2011 Institut Teknologi Bandung       
       Fax : +62-22-250 4010, +62-22-251 1215 XHTML + CSS + RSS       
       E-mail : or Developed by AVE