Learning about Voice Search for Spoken Dialogue Systems

In a Wizard-of-Oz experiment with multiple wizard subjects, each wizard viewed automated speech recognition (ASR) results for utterances whose interpretation is critical to task success: requests for books by title from a library database. To avoid non-understandings, the wizard directly queried the application database with the ASR hypothesis (voice search). To learn how to avoid misunderstandings, we investigated how wizards dealt with uncertainty in voice search results. Wizards were quite successful at selecting the correct title from query results that included a match. The most successful wizard could also tell when the query results did not contain the requested title. Our learned models of the best wizard's behavior combine features available to wizards with some that are not, such as recognition confidence and acoustic model scores.

• Type: Journal
• Year: 2010
• Where: NAACL
• Authors: Rebecca J. Passonneau, Susan L. Epstein, Tiziana Ligorio, Joshua B. Gordon, Pravin Bhutada
• Keywords: Acoustic Model Scores, Computational Linguistics, Multiple Wizard Subjects, NAACL 2010, Voice Search