Porting Decipher from English to Mandarin
Mei-Yuh Hwang, Xin Lei, T. Ng, Mari Ostendorf, Andreas Stolcke, W. Wang, Jing Wen Zheng, V.R.R. Gadde · 2006
This paper describes our efforts in porting the SRI Decipher English system into Mandarin for transcribing tele-phone conversations. This includes all aspects of the system: the pronunciation phone set and lexicon, word segmen-tation, pitch features, discriminatively trained acoustic models with parameter sharing determined by decision trees, and web-data augmented language models.