You are here

Learning Hidden Unit Contributions for Unsupervised Speaker Adaptation of Neural Network Acoustic Models

Title Learning Hidden Unit Contributions for Unsupervised Speaker Adaptation of Neural Network Acoustic Models
Publication Type Conference Paper
2014
Authors Swietojanski, P, Renals, S
Conference Name Proc. IEEE Workshop on Spoken Language Technology
Date Published December
Conference Location Lake Tahoe, USA

This paper proposes a simple yet effective model-based neural network speaker adaptation technique that learns speaker- specific hidden unit contributions given adaptation data, without requiring any form of speaker-adaptive training, or labelled adaptation data. An additional amplitude parameter is defined for each hidden unit; the amplitude parameters are tied for each speaker, and are learned using unsupervised adaptation. We conducted experiments on the TED talks data, as used in the International Workshop on Spoken Language Translation (IWSLT) evaluations. Our results indicate that the approach can reduce word error rates on standard IWSLT test sets by about 8–15% relative compared to unadapted systems, with a further reduction of 4–6% relative when combined with feature-space maximum likelihood linear re- gression (fMLLR). The approach can be employed in most existing feed-forward neural network architectures, and we report results using various hidden unit activation functions: sigmoid, maxout, and rectifying linear units (ReLU).