Volume 23, Issue 2 (1-2005)
2005, 23(2): 39-50 |
Back to browse issues page
Download citation:
BibTeX | RIS | EndNote | Medlars | ProCite | Reference Manager | RefWorks
Send citation to:
BibTeX | RIS | EndNote | Medlars | ProCite | Reference Manager | RefWorks
Send citation to:
S. Sharifian and S. M. Ahadi. Speaker Adaptation in Continuous Speech Recognition Using MLLR-Based MAP Estimation. Computational Methods in Engineering 2005; 23 (2) :39-50
URL: http://jcme.iut.ac.ir/article-1-317-en.html
URL: http://jcme.iut.ac.ir/article-1-317-en.html
Abstract: (3816 Views)
A variety of methods are used for speaker adaptation in speech recognition. In some techniques, such as MAP estimation, only the models with available training data are updated. Hence, large amounts of training data are required in order to have significant recognition improvements. In some others, such as MLLR, where several general transformations are applied to model clusters, the results are desirable for small training data, but with increasing training data, the performance improvement reaches the saturation lvel. In this paper, a new approach is introduced that makes use of the advantages of both mentioned techniques to improve the recognition rate. Here, the models with available training data are trained using MAP while
for those with insufficient training data, appropriate prior parameters for MAP estimation are found using MLLR. This technique has yielded better performance in comparison to either MAP or MLLR, in a system based on FARSDAT speech corpus.
Keywords: Hidden Markov models (HMM), continuous Persian (Farsi) speech recognition, Speaker adaptation, MAP estimation, MLLR transformation.
