Reordering: A stepping-stone to perfect Thai sign generation

• Main Authors: Srisavakon Dangsaart, Nick Cercone, Kanlaya Naruedomkul, Booncharoen Sirinaovakul
• Other Authors: Mahidol University
• Format: Conference or Workshop Item
• Published: 2018
• Subjects: Computer Science; Engineering
• Online Access: https://repository.li.mahidol.ac.th/handle/123456789/24391

We proposed the Sign Code Reordering approach (SCR) for reordering the intermediate sign codes (ISC) to Sign code script (SCS) generation. SCR uses language structure matching techniques to reduce complicated grammar rules, provide efficient results. SCR comprises three steps: extraction, reordering and integration. The distinction between source and target language in both grammar and vocabulary is concerned in each processing step to ensure the accuracy of reordering. SCR focuses on accurate and acceptable reordering that are not conforming to the original structure. SCR was designed to capture linguistic differences such as phrase, sentence and multi-sentence structures, no matter how long the input sentence is. The SCR prototype system was implemented and tested to reorder a number ISCs. The test results have been proved that SCR arranges ISCs successfully. SCR can be augmented into any NLP application which requires ISC arrangement e.g., T3STS. T3STS translates Thai text into Thai Sign language. Thai Sign language is the language of the Deaf in Thailand.