tailieunhanh - Báo cáo khoa học: "A Shallow Model of Backchannel Continuers in Spoken Dialogue"

Spoken dialogue systems would be more acceptable if they were able to produce backchannel continuers such as mm-hmm in naturalistic locations during the user's utterances. Using the HCRC Map Task Corpus as our data source, we describe models for predicting these locations using only limited processing and features of the user's speech that are commonly available, and which therefore could be used as a lowcost improvement for current systems. The baseline model inserts continuers after a predetermined number of words. . | A Shallow Model of Backchannel Continuers in Spoken Dialogue Nicola Cathcart Canon Research Centre Europe Bracknell UK nicolac@ Jean Carletta and Ewan Klein School of Informatics University of Edinburgh jeanc ewan @ Abstract Spoken dialogue systems would be more acceptable if they were able to produce backchannel continuers such as mm-hmm in naturalistic locations during the user s utterances. Using the HCRC Map Task Corpus as our data source we describe models for predicting these locations using only limited processing and features of the user s speech that are commonly available and which therefore could be used as a lowcost improvement for current systems. The baseline model inserts continuers after a predetermined number of words. One further model correlates back-channel continuers with pause duration while a second predicts their occuưence using trigram POS frequencies. Combining these two models gives the best results. 1 Introduction In a spoken dialogue between people the participants use simple utterances such as yeah a totty wee bit aye and mm-hmm to signal that communication is working. Without this feedback the partner may assume that he has not been understood and reformulate his utterance. Following Yngve 1970 we will use the term backchannel for such utterances. Although these can be substantive because they can repeat material from the partner s utterance Clark and Schaefer 1991 . Right okay I m below the flat rocks we will adopt Ju-rafsky et al. 1998 s terminology of continuer. We will take this to refer to the class of backchannel utterances with minimal content used to clearly signal that the speaker should continue with her current turn. Yankelovich et al. 1995 point out that users of speech interface systems need feedback too especially since the system s silence could mean either of two very different things that it is waiting for user input in which case the user should speak or that it is still processing .