Shaping user input in speech graffiti: a first pass

Speech Graffiti is a standardized interaction protocol for spoken dialog systems designed to address some com-mon difficulties with ASR. We have proposed a strategy of shaping to help users adapt their interaction to match what the system understands best, thereby re-ducing the chance for misunderstandings and improving interaction efficiency. In this paper we report on an evaluation of our initial implementation of shaping in Speech Graffiti, noting that our baseline strategy was not as powerful as expected, and discussing proposed changes to improve its effectiveness.