We report results for an online multi-keyword spotter in a game that contains overlapping speech, off-task side talk, and keyword forms that vary in completeness and duration. The spotter trained on a dataset of 62 children, and expectations for online performance were established by 10-fold cross-validation on that corpus. We compare the post hoc data to the recognizer's performance online in a study in which 24 new children played with the real-time system. The online system showed a non-significant decline in accuracy which could be traced to trouble understanding the jump keyword and the predominance of younger children in the new cohort. However, children adjusted their behavior to compensate, and the overall performance and responsiveness of the online system resulted in engaging and enjoyable gameplay.
Cite as: Lehman, J.F., Wolfe, N., Pereira, A. (2016) G-g-go! Juuump! Online Performance of a Multi-keyword Spotter in a Real-time Game. Proc. 5th Workshop on Child Computer Interaction (WOCCI 2016), 51-55, doi: 10.21437/WOCCI.2016-9
@inproceedings{lehman16_wocci,
author={Jill Fain Lehman and Nikolas Wolfe and André Pereira},
title={{G-g-go! Juuump! Online Performance of a Multi-keyword Spotter in a Real-time Game}},
year=2016,
booktitle={Proc. 5th Workshop on Child Computer Interaction (WOCCI 2016)},
pages={51--55},
doi={10.21437/WOCCI.2016-9}
}