Mobile phones, with their increasing processing power and memory, are enabling a diversity of tasks. The traditional text entry method using keypad is falling short in numerous way...
Santosh Basapur, Shuang Xu, Mark Ahlenius, Young S...
State-of-the-art large vocabulary continuous speech recognition (LVCSR) systems often combine outputs from multiple subsystems developed at different sites. Cross system adaptatio...
In this paper we present an overview on the development of a large vocabulary continuous speech recognition (LVCSR) system for Khmer, the official language of Cambodia, spoken by ...
Long-span language models that capture syntax and semantics are seldom used in the first pass of large vocabulary continuous speech recognition systems due to the prohibitive sea...
Anoop Deoras, Tomas Mikolov, Stefan Kombrink, Mart...
In discriminative training, such as Maximum Mutual Information Estimation (MMIE) training, a word lattice is usually used as a compact representation of many different sentence hy...