Publications
Detail of publication
Citation
p. 212-217, Západočeská univerzita, Plzeň, 2007. : The Speaker Adaptation of an Acoustic Model . The 1st Young Researchers Conference on Applied Sciences,
Download PDF
Abstract
This paper deals with several adaptation techniques, which are of the importance in cases when the identity of a speaker is known and we want to recognize his speech. Each of the methods yields various benefits, therefore we examined their combination. This approach brought further error rate decreasing. We are using three different methods, namely Maximum Apriori Probability adaptation, Maximum Likelihood Linear Regression and Constrained Maximum Likelihood Linear Regression. All acoustic models are based on the Hiden Markov Model.
Detail of publication
Title: | The Speaker Adaptation of an Acoustic Model |
---|---|
Author: | Machlica, L. ; Zajíc, Z. |
Language: | English |
Date of publication: | 13 Nov 2007 |
Year: | 2007 |
Type of publication: | Papers in proceedings of reviewed conferences |
Title of journal or book: | The 1st Young Researchers Conference on Applied Sciences |
Page: | 212 - 217 |
ISBN: | 978-80-7043-574-8 |
Publisher: | Západočeská univerzita |
Address: | Plzeň |
Date: | 13 Nov 2007 - 13 Nov 2007 |
Keywords
adaptation, MAP, MLLR, CMLLR, combination
BibTeX
@INPROCEEDINGS{MachlicaL_2007_TheSpeaker, author = {Machlica, L. and Zaj\'{i}c, Z.}, title = {The Speaker Adaptation of an Acoustic Model}, year = {2007}, publisher = {Z\'{a}pado\v{c}esk\'{a} univerzita}, journal = {The 1st Young Researchers Conference on Applied Sciences}, address = {Plze\v{n}}, pages = {212-217}, ISBN = {978-80-7043-574-8}, url = {http://www.kky.zcu.cz/en/publications/MachlicaL_2007_TheSpeaker}, }