Understanding speech in interactive narratives with crowd sourced data

Understanding speech in interactive narratives with crowd sourced data

Speech recognition failures and limited vocabulary coverage pose challenges for speech interactions with characters in games. We describe an end-to-end system for automating characters from a large corpus of recorded human game logs, and demonstrate that inferring utterance meaning through a combina...

Full description

Bibliographic Details
Main Authors: Orkin, Jeffrey David (Contributor), Roy, Deb K. (Contributor)
Other Authors: Massachusetts Institute of Technology. Media Laboratory (Contributor), Program in Media Arts and Sciences (Massachusetts Institute of Technology) (Contributor)
Format: Article
Language:English
Published: Association for the Advancement of Artificial Intelligence (AAAI), 2014-12-23T15:16:40Z.
Subjects:
Article
Online Access:Get fulltext
Description
Summary:Speech recognition failures and limited vocabulary coverage pose challenges for speech interactions with characters in games. We describe an end-to-end system for automating characters from a large corpus of recorded human game logs, and demonstrate that inferring utterance meaning through a combination of plan recognition and surface texts similarity compensates for recognition and understanding failures significantly better than relying on surface similarity alone.
Singapore-MIT GAMBIT Game Lab

Similar Items