|Date||Nov 21, 2013|
|Title||Automatic transcription system for the Japanese Parliament (Diet)|
Applications of automatic speech recognition (ASR) have been extended toa variety of tasks and domains, including spontaneous human-human speech. We have developed an ASR system for theJapanese Parliament (Diet), which has been operated since 2011. By exploiting official records made by humanstenographers, we have realized an efficient training scheme of acoustic andlanguage models, which does not require faithful transcripts and thus isscalable to enormous data. This talk will give an overview of the system andalso an ASR system for lectures investigated in Kyoto University.
Tatsuya Kawahara received B.E. in 1987, M.E. in 1989, and Ph.D. in 1995,all in information science, from Kyoto University, Kyoto, Japan. Currently, heis a Professor in the Academic Center for Computing and Media Studies and anAffiliated Professor in the School of Informatics, Kyoto University. He hasalso been an Invited Researcher at ATR and NICT.
He has published more than 250 technical papers on speech recognition, spokenlanguage processing, and spoken dialogue systems. He has been conductingseveral speech-related projects in Japan including free large vocabularycontinuous speech recognition software (http://julius.sourceforge.jp/) and theautomatic transcription system for the Japanese Parliament (Diet).
Dr. Kawahara received the Commendation for Science and Technology by theMinister of Education, Culture, Sports, Science and Technology (MEXT) in 2012.From 2003 to 2006, he was a member of IEEE SPS Speech Technical Committee. He was a general chair of IEEE AutomaticSpeech Recognition \& Understanding workshop (ASRU 2007). He also served asa Tutorial Chair of INTERSPEECH 2010 and a Local Arrangement Chair of IEEEICASSP 2012. He is an editorial board member of Elsevier Journal of ComputerSpeech and Language, ACM Transactions on Speech and Language Processing, andAPSIPA Transactions on Signal and Information. He is a senior member of IEEE.
Please note that this seminar series talk will not be held onFriday as usual. This talk will be held at 2 p.m. in the IMR onThursday.