Acoustic Adaptation Using Non-Linear Transformations of HMM Parameters


Abrash, V., Sankar, A., Franco, H., & Cohen, M. (1996, May). Acoustic adaptation using nonlinear transformations of HMM parameters. In 1996 IEEE International Conference on Acoustics, Speech, and Signal Processing Conference Proceedings (Vol. 2, pp. 729-732). IEEE.


Speech recognition performance degrades significantly when there is a mismatch between testing and training conditions. Linear transformation-based maximum-likelihood (ML) techniques have been proposed recently to tackle this problem. In this paper, we extend this approach to use nonlinear transformations. These are implemented by multilayer perceptrons (MLPs) which transform the Gaussian means. We derive a generalized expectationmaximization (GEM) training algorithm to estimate the MLP weights. Some preliminary experimental results on nonnative speaker adaptation are presented.

Read more from SRI