7th International Conference on Spoken Language Processing

September 16-20, 2002
Denver, Colorado, USA

Can Confidence Scores Help Users Post-Editing Speech Recognizer Output?

Taku Endo (1), Nigel Ward (2), Minoru Terada (1)

(1) University of Tokyo, Japan; (2) University of Texas at El Paso, USA

When dictating with speech recognition, most of the userís time is spent correcting errors. To decrease the burden we propose new editor functions specifically to speed up the correction process. The idea is to use a recognition confidence measure to predict which words are likely to be in error, to display that information to the user by highlighting suspect words, and to provide a command to let the user jump the cursor to the next suspect word. Simple experiments suggest that these functions can be valuable, even with todayís speech recognizers and confidence measures.


Full Paper

Bibliographic reference.  Endo, Taku / Ward, Nigel / Terada, Minoru (2002): "Can confidence scores help users post-editing speech recognizer output?", In ICSLP-2002, 1469-1472.