The HCRC Map Task Corpus: Natural Dialogue for Speech Recognition

THE HCRC MAP TASK CORPUS: NATURAL DIALOGUE FOR SPEECH RECOGNITION Henry S. Thompson 1,2,3 Anne Anderson1, 5 Ellen Gurman Bard 1,3,4 Gwyneth Doherty-Sneddon l, 5 Alison Newlands l, 5 Cathy Sotillo 1,4 1: Human Communication Research Centre 2: Department of Artificial Intelligence 3: Centre for Cognitive Science 4: Department of Linguistics University of Edinburgh 2 Buccleuch Place, Edinburgh, EH12 5BB SCOTLAND 5: Department of Psychology University of Glasgow 56 t-lillhead Street Glasgow, G12 SCOTLAND hthompson@edinburgh.ac.uk ABSTRACT The HCRC Map Task corpus has been collected and transcribed in Glasgow and Edinburgh, and recently published on CD-ROM. This effort was made possible by funding from the British Economic and Social Research Council. The corpus is composed of 128 two-person conver- sations in both high-quality digital audio and orthographic transcriptions, amounting to 18 hours and 150,000 words respectively. The experimental design is quite detailed and complex, allowing a number of different phone- mic, syntactico-semantic and pragmatic con- trasts to be explored in a controlled way. The corpus is a uniquely valuable resource for speech recognition research in particular, as we move from developing systems intended for con- trolled use by familiar users to systems intended for less constrained circumstances and naive or occasional users. Examples supporting this claim are given, including preliminary evi- dence of the phonetic consequences of second mention and the impact of different styles of ref- erent negotiation on communicative efficacy. 1. INTRODUCTION The HCRC Map Task corpus has been col- lected and transcribed in Glasgow and Edinburgh, and recently published on CD- ROM (HCRC 1993). This effort was made possible by funding from the British Economic and Social Research Council. The group which designed and collected the corpus covers a wide range of interests and the corpus reflects this, providing a resource for studies of natural dialogue from many different perspectives. In this paper we will give a brief summary of the experimental design, and then concen- trate on those aspects of the corpus which make it a uniquely valuable resource for speech recognition research in particular, as we move from developing systems intended 25 for controlled use by familiar users to sys- tems intended for less constrained circum- stances and naive or occasional users. Some preliminary results of work on the phonetic consequences of second mention and on the impact of different styles of referent negotia- tion on communicative efficacy will also be presented. 2. CORPUS DESIGN AND CHARACTERISTICS 2.1. The Task The conversations were elicited by an exer- cise in task-oriented cooperative problem solving. The two participants sat facing one another in a small recording studio, sepa- rated by a table on which sat back-to-back reading stands. On each stand was a schematic map, each visible only to one par- ticipant. Each map consisted of an outline and roughly a dozen labelled features (e.g. "white cottage", "Green Bay", "oak forest"). Most features are common to the two maps, but not all, and the participants were in- formed of this. One map had a route drawn in, the …

Content truncated to 500 words