Investigation of Maxout Networks For Speech Recognition

Pawel Swietojanski, Jinyu Li, and Jui-Ting Huang

Abstract

We explore the use of maxout neuron in various aspects of acoustic

modelling for large vocabulary speech recognition systems; including

low-resource scenario and multilingual knowledge transfers.

Through the experiments on voice search and short message dictation

datasets, we found that maxout networks are around three times

faster to train and offer lower or comparable word error rates on

several tasks, when compared to the networks with logistic nonlinearity.

We also present a detailed study of the maxout unit internal

behaviour suggesting the use of different nonlinearities in different

layers.

Details

Publication typeProceedings
Published inICASSP
> Publications > Investigation of Maxout Networks For Speech Recognition