作者: T. Zeppenfeld , E. Shriberg , M. Ostendorf , M. Finke , S. Roweis
DOI:
关键词:
摘要: This paper describes the research efforts of “Hidden Speaking Mode” group participating in 1996 summer workshop on speech recognition. The goal this project is to model pronunciation variations that occur conversational general and, more specifically, investigate use a hidden speaking mode represent systematic are correlated with word sequence (e.g. predictable from syntactic structure). theoretical formulation modeling, as well some results error analysis, language modeling and modeling.