A Tale of Two Tasks: Detecting Children's Off-Task Speech in a Reading Tutor - Robotics Institute Carnegie Mellon University

A Tale of Two Tasks: Detecting Children’s Off-Task Speech in a Reading Tutor

Wei Chen and Jack Mostow
Conference Paper, Proceedings of 12th Annual Conference of the International Speech Communication Association (INTERSPEECH '11), pp. 1621 - 1624, August, 2011

Abstract

How can an automated tutor detect children’s off-task utterances? To answer this question, we trained SVM classifiers on a corpus of 495 children’s 36,492 computerassisted oral reading utterances. On a test set of 620 utterances by 10 held-out readers, the classifier correctly detected 88% of off-task utterances and misclassified 17% of on-task utterances as off-task. As a test of generality, we applied the same classifier to 20 children’s 410 responses to vocabulary questions. The classifier detected 84% of off-task utterances but misclassified 57% of on-task utterances. Acoustic and lexical features helped detect off-task speech in both tasks.

BibTeX

@conference{Chen-2011-122079,
author = {Wei Chen and Jack Mostow},
title = {A Tale of Two Tasks: Detecting Children's Off-Task Speech in a Reading Tutor},
booktitle = {Proceedings of 12th Annual Conference of the International Speech Communication Association (INTERSPEECH '11)},
year = {2011},
month = {August},
pages = {1621 - 1624},
}