Recognition and Correction of Voice Web Search Queries

Keith Vertanen, Per Ola Kristensson

Interspeech '09: Proceedings of the International Conference on Spoken Language Processing, 2009.

In this work we investigate how to recognize and correct voice web search queries. We describe our corpus of web search queries and show how it was used to improve recognition accuracy. We show that using a search-specific vocabulary with automatically generated pronunciations is superior to using a vocabulary limited to a fixed pronunciation dictionary. We conducted a formative user study to investigate recognition and correction aspects of voice search in a mobile context. In the user study, we found that despite a word error rate of 48%, users were able to speak and correct search queries in about 18 seconds. Users did this while walking around using a mobile touch-screen device.

There are more details about this research in chapter 6 of my thesis.