For Full-Text PDF, please login, if you are a member of IEICE,|
or go to Pay Per View on menu list, if you are a nonmember of IEICE.
Speaker Adaptation Based on PPCA of Acoustic Models in a Two-Way Array Representation
IEICE TRANSACTIONS on Information and Systems
Publication Date: 2014/08/01
Online ISSN: 1745-1361
Type of Manuscript: LETTER
Category: Speech and Hearing
expectation-maximization, probabilistic principal component analysis, speaker adaptation, speech recognition,
Full Text: PDF(90.3KB)>>
We propose a speaker adaptation method based on the probabilistic principal component analysis (PPCA) of acoustic models. We define a training matrix which is represented in a two-way array and decompose the training models by PPCA to construct bases. In the two-way array representation, each training model is represented as a matrix and the columns of each training matrix are treated as training vectors. We formulate the adaptation equation in the maximum a posteriori (MAP) framework using the bases and the prior.