Paper: Detecting Retries of Voice Search Queries

ACL ID P14-2038
Title Detecting Retries of Voice Search Queries
Venue Annual Meeting of the Association of Computational Linguistics
Session Main Conference
Year 2014
Authors

When a system fails to correctly recog- nize a voice search query, the user will fre- quently retry the query, either by repeat- ing it exactly or rephrasing it in an attempt to adapt to the system?s failure. It is de- sirable to be able to identify queries as retries both offline, as a valuable quality signal, and online, as contextual informa- tion that can aid recognition. We present a method than can identify retries offline with 81% accuracy using similarity mea- sures between two subsequent queries as well as system and user signals of recogni- tion accuracy. The retry rate predicted by this method correlates significantly with a gold standard measure of accuracy, sug- gesting that it may be useful as an offline predictor of accuracy.