Title :
A posterior union model for robust speech recognition
Author :
Ming, Ji ; Lin, Jie
Author_Institution :
Dept. of Comput. Sci., Queen´´s Univ., Belfast, UK
Abstract :
The probabilistic union model is extended from a conditional-probability formulation to a posterior-probability formulation for speech recognition. The new formulation allows the order of the model to be optimised for every single frame, thereby enhancing the capability of the model for dealing with nonstationary noise corruption.
Keywords :
hidden Markov models; probability; speech recognition; HMM; nonstationary noise corruption; posterior union model; posterior-probability formulation; robust speech recognition;
Journal_Title :
Electronics Letters
DOI :
10.1049/el:20030016
