ISCA Archive SLaTE 2011
ISCA Archive SLaTE 2011

Evaluating tracking accuracy of an automatic reading tutor

Morten Højfeldt Rasmussen, Jack Mostow, Zheng-Hua Tan, Børge Lindberg, Yuanpeng Li

In automatic reading tutoring, tracking is the process of automatically following a reader through a given target text. When developing tracking algorithms, a measure of the tracking accuracy – how often a spoken word is aligned to the right target text word position – is needed in order to evaluate performance and compare different algorithms. This paper presents a framework for determining the observed tracking error rate. The proposed framework is used to evaluate three tracking strategies: A) follow the reader to whichever word he/she jumps to in the text, B) follow the reader monotonically from left to right ignoring word skips and regressions (going back to a previous text word), and C) the same as B but allowing isolated word skips. Observed tracking error rate for each of the three tracking strategies is: A: 53%, B: 56%, and C: 47%, on 1883 utterances from 25 children.

Index Terms. Automatic Reading Tutor, Tracking Speech, Tracking Error Rate


Cite as: Rasmussen, M.H., Mostow, J., Tan, Z.-H., Lindberg, B., Li, Y. (2011) Evaluating tracking accuracy of an automatic reading tutor. Proc. Speech and Language Technology in Education (SLaTE 2011), 17-20

@inproceedings{rasmussen11_slate,
  author={Morten Højfeldt Rasmussen and Jack Mostow and Zheng-Hua Tan and Børge Lindberg and Yuanpeng Li},
  title={{Evaluating tracking accuracy of an automatic reading tutor}},
  year=2011,
  booktitle={Proc. Speech and Language Technology in Education (SLaTE 2011)},
  pages={17--20}
}