The Games Project is a long-term research effort involving the
collection and prosodic analysis of a multi-speaker corpus of
spontaneous dialogue. Although originally conceived to address specific
hypotheses on how intonation varies in relation to information status
and discourse structure, it is also proving to be useful for studying
many other open speech-related questions. Currently, we are looking at
two particular questions in spoken dialogue systems: speakers' use of
discourse markers (such as 'okay', 'uh-huh' or 'yeah'), and turn-taking
strategies. In this talk, I will comment on the design and original
goals of our research, describe the current state of our investigations,
and outline future research.
|