Automatic Detection and Correction of Repairs in Human-Computer Dialog

Automatic Detection and Correction of Repairs in Human-Computer Dialog* Elizabeth Shriberg~ John Bear, John Doweling SRI International Menlo Park, California 94025 ABSTRACT We have analyzed 607 sentences of spontaneous human- computer speech data containing repairs (drawn from a cor- pus of 10,718). We present here criteria and techniques for automatically detecting the presence of a repair, its loca- tion, and making the appropriate correction. The criteria in- volve integration of knowledge from several sources: pattern matching, syntactic and semantic analysis, and acoustics. 1. INTRODUCTION Spontaneous spoken language often includes speech that is not intended by the speaker to be part of the con- tent of the utterance. This speech must be detected and deleted in order to correctly identify the intended meaning. This broad class of disfluencies encompasses a number of phenomena, including word fragments, in- terjections, filled pauses, restarts, and repairs. We are analyzing the repairs in a large subset (over ten thou- sand sentences) of spontaneous speech data collected for the DARPA spoken language program. We have cate- gorized these disfluencies as to type and frequency, and are investigating methods for their automatic detection and correction. Here we report promising results on de- tection and correction of repairs by combining pattern matching, syntactic and semantic analysis, and acous- tics. The problem of disfluent speech for language under- standing systems has been noted but has received limited attention, ttindle [5] attempts to delimit and correct repairs in spontaneous human-human dialog, based on transcripts containing an "edit signal," or external and reliable marker at the "expunction point," or point of in- terruption. Carbonell and Hayes [4] briefly describe re- *This research was supported by the Defense Advanced Re- search Projects Agency under Contract ONR N00014-90-C-0085 with the Office of Naval Research. It was also supported by a Grant, NSF IRI-8905249, from the NationM Science Foundation. The views and conclusions contained in this document are those of the authors and should not be interpreted as necessarily represent- ing the official policies, either expressed or implied, of the Defense Advanced Research Projects Agency of the U.S. Government, or of the National Science Foundation. lElizabeth Shriberg is also affiliated with the Department of Psychology at the University of California at Berkeley. covery strategies for broken-off and restarted utterances in textual input. Ward [13] addresses repairs in sponta- neous speech, but does not attempt to identify or correct them. Our approach is most similar to that of Hindle. It differs, however, in that we make no assumption about the existence of an explicit edit signal. As a reliable edit signal has yet to be found, we take it as our problem to find the site of the repair automatically. It is the case, however, that cues to repair exist over a range of syllables. Research in speech production has shown that repairs tend to be marked prosodically [8] and there is perceptual evidence from work using lowpass-filtered speech that human listeners can detect the occurrence of a repair in the absence of segmental information …

Content truncated to 500 words