Title :
Adaptation method using expectation-maximisation for noisy speech recognition
Author_Institution :
Dept. of Electron., Keimyung Univ., Daegu, South Korea
fDate :
6/20/2002 12:00:00 AM
Abstract :
An expectation-maximisation-based adaptation method for the mean of noise is proposed for noisy speech recognition. The noisy testing speech is used for unsupervised adaptation by employing the parallel model combination (PMC). The proposed method has shown improved performances compared with the PMC method in isolated word recognition experiments
Keywords :
acoustic noise; iterative methods; speech recognition; adaptation method; expectation-maximisation; noisy speech recognition; parallel model combination; unsupervised adaptation; word recognition;
Journal_Title :
Electronics Letters
DOI :
10.1049/el:20020428
