Improving ASR performance using context-dependent phoneme model

• Main Authors: Husni, Husniza, Jamaludin, Zulikha
• Format: Article
• Language: English
• Published: Emerald Group Publishing 2010
• Subjects: LC Special aspects of education; QA75 Electronic computers. Computer science
• Online Access: http://repo.uum.edu.my/1319/1/Zulikha%2CJ_Improving_ASR%5B1%5D.pdf; http://repo.uum.edu.my/1319/; http://www.emeraldinsight.com/1328-7265.htm

Purpose – The purpose of this paper is to present evidence of the need to have a carefully designed lexical model for speech recognition for dyslexic children reading in Bahasa Melayu (BM). Design/methodology/approach – Data collection is performed to obtain the most frequent reading error patterns and the reading recordings. Design and development of the lexical model considers the errors for better recognition accuracy. Findings – It is found that the recognition accuracy is increased to 75 percent when using contextdependent (CD) phoneme model and phoneme refinement rule. Comparison between contextindependent phoneme models and CD phoneme model is also presented. Research limitations/implications – The most frequent errors recognized and obtained from data collection and analysis illustrate and support that phonological deficit is the major factor for reading disabilities in dyslexics. Practical implications – This paper provides the first step towards materializing an automated speech recognition (ASR)-based application to support reading for BM, which is the first language in Malaysia. Originality/value – The paper contributes to the knowledge of the most frequent error patterns for dyslexic children’s reading in BM and to the knowledge that a CD phoneme model together with the phoneme refinement rule can built up a more fine-tuned lexical model for an ASR specifically for dyslexic children’s reading isolated words in BM.