Towards Machine Comprehension of Spoken Content: Initial TOEFL Listening Comprehension Test by Machine

TL;DR

This paper introduces a new task of machine comprehension of spoken content, specifically focusing on TOEFL listening tests, and proposes an Attention-based Multi-hop Recurrent Neural Network to address it, showing promising initial results.

Contribution

The paper defines a novel task of machine comprehension for spoken content and develops an AMRNN model tailored for this challenge, incorporating attention mechanisms.

Findings

Word-level attention outperforms sentence-level attention with ASR errors.

The proposed AMRNN achieves encouraging initial results on TOEFL listening comprehension.

Spoken content comprehension remains a challenging but promising research area.

Abstract

Multimedia or spoken content presents more attractive information than plain text content, but it's more difficult to display on a screen and be selected by a user. As a result, accessing large collections of the former is much more difficult and time-consuming than the latter for humans. It's highly attractive to develop a machine which can automatically understand spoken content and summarize the key information for humans to browse over. In this endeavor, we propose a new task of machine comprehension of spoken content. We define the initial goal as the listening comprehension test of TOEFL, a challenging academic English examination for English learners whose native language is not English. We further propose an Attention-based Multi-hop Recurrent Neural Network (AMRNN) architecture for this task, achieving encouraging results in the initial tests. Initial results also have shown that word-level attention is probably more robust than sentence-level attention for this task with ASR errors.