Multi-Attribute Factorized Hidden Layer Adaptation for DNN Acoustic Models

Lahiru Samarakoon, Khe Chai Sim


Recently, the Factorized Hidden Layer (FHL) adaptation is proposed for speaker adaptation of deep neural network (DNN) based acoustic models. In addition to the standard affine transformation, an FHL contains a speaker-dependent (SD) transformation matrix using a linear combination of rank-1 matrices and an SD bias using a linear combination of vectors. In this work, we extend the FHL based adaptation to multiple variabilities of the speech signal. Experimental results on Aurora4 task show 26.0% relative improvement over the baseline when standard FHL adaptation is used for speaker adaptation. The Multi-attribute FHL adaptation shows gains over the standard FHL adaptation where improvements reach up to 29.0% relative to the baseline.


DOI: 10.21437/Interspeech.2016-1233

Cite as

Samarakoon, L., Sim, K.C. (2016) Multi-Attribute Factorized Hidden Layer Adaptation for DNN Acoustic Models. Proc. Interspeech 2016, 3484-3488.

Bibtex
@inproceedings{Samarakoon+2016,
author={Lahiru Samarakoon and Khe Chai Sim},
title={Multi-Attribute Factorized Hidden Layer Adaptation for DNN Acoustic Models},
year=2016,
booktitle={Interspeech 2016},
doi={10.21437/Interspeech.2016-1233},
url={http://dx.doi.org/10.21437/Interspeech.2016-1233},
pages={3484--3488}
}