{"title":"Comparison among Various Question Generations for Decision Tree Based State Tying in Persian Language","authors":"Nasibeh Nasiri, Dawood Talebi Khanmiri","volume":62,"journal":"International Journal of Computer and Information Engineering","pagesStart":197,"pagesEnd":201,"ISSN":"1307-6892","URL":"https:\/\/publications.waset.org\/pdf\/11978","abstract":"<p>Performance of any continuous speech recognition system is highly dependent on performance of the acoustic models. Generally, development of the robust spoken language technology relies on the availability of large amounts of data. Common way to cope with little data for training each state of Markov models is treebased state tying. This tying method applies contextual questions to tie states. Manual procedure for question generation suffers from human errors and is time consuming. Various automatically generated questions are used to construct decision tree. There are three approaches to generate questions to construct HMMs based on decision tree. One approach is based on misrecognized phonemes, another approach basically uses feature table and the other is based on state distributions corresponding to context-independent subword units. In this paper, all these methods of automatic question generation are applied to the decision tree on FARSDAT corpus in Persian language and their results are compared with those of manually generated questions. The results show that automatically generated questions yield much better results and can replace manually generated questions in Persian language.<\/p>\r\n","references":"[1] K. Beulen, H. Ney, \"Automatic question generation for decision tree based state tying,\" Proc. of ICASSP '98, pp. 805-808, 12-15 May, Seattle, WA, USA, 1998.\r\n[2] J. J. Odell, \"The Use of Context in Large Vocabulary Speech Recognition,\" Ph.D. Thesis, Cambridge University, 1995.\r\n[3] M. Bijankhan et al., \"FARSDAT - The Speech Database of Farsi Spoken Language\", Proc. 5th Australian Int. Conf. On Speech Science and Tech., Vol. 2, perth, 1994.\r\n[4] Singh, R., Raj, B., Stern, R. M.: Automatic Clustering and Generation of\r\nContextual Questions for Tied States in Hidden Markov Models. In Proc.\r\nICSLP, Vol. 1, pp.117-1202, 1999\r\n[5] Kanokphara, S., Geumann, A.,Carson-Berndsen, J.: Accessing Language\r\nSpecific Linguistic Information for Triphone Model Generation: Feature\r\nTables in a Speech Recognition System., 2nd Language & Technology\r\nConference: Human Language Technologies as a Challenge for Computer\r\nScience and Linguistics, 2005.\r\n[6] Kanokphara, S. and Carson-Berndsen, J.: Automatic Question Generation\r\nfor HMM State Tying using a Feature Table. Proc. Australian Int. Conf. on\r\nSpeech Science & Technology (ASST) 2004.\r\n[7] Kanokphara. S. , Carson-Berndsen, J.: Phonetic Question Generation Using Misrecongnition,\", In Proc. The Ninth Inernational Conference on\r\nTEXT , SPEECH and DIALOGE(TSD), Brno, Czech Republic,\r\nSeptember, pp. 407-414, 2006.","publisher":"World Academy of Science, Engineering and Technology","index":"Open Science Index 62, 2012"}