Align or attend?

Toward More Efficient and Accurate Spoken Word Discovery Using Speech-to-Image Retrieval

Conference paper (2021)

Authors

Liming Wang University of Illinois at Urbana Champaign

X. Wang Xi’an Jiaotong University,

Mark Hasegawa-Johnson University of Illinois at Urbana Champaign

O.E. Scharenborg

Najim Dehak Johns Hopkins University

Language acquisition Low-resource speech technology Multimodal learning Spoken term discovery

To reference this document use:

http://resolver.tudelft.nl/uuid:043e055f-194b-4f66-a399-8e76e141429a

More Info

expand_more

Published Date

2021

Language

English

Reuse Rights

Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons.

Abstract

Multimodal word discovery (MWD) is often treated as a byproduct of the speech-to-image retrieval problem. However, our theoretical analysis shows that some kind of alignment/attention mechanism is crucial for a MWD system to learn meaningful word-level representation. We verify our theory by conducting retrieval and word discovery experiments on MSCOCO and Flickr8k, and empirically demonstrate that both neural MT with self-attention and statistical MT achieve word discovery scores that are superior to those of a state-of-the-art neural retrieval system, outperforming it by 2% and 5% alignment F1 scores respectively.

Files

09414418.pdf

(pdf | 3.17 Mb)

Unknown license

Download not available