A SPOKEN DIALOG SYSTEM WITH ACTIVE NON-ACTIVE WORD CONTROL FOR CD-ROMINFORMATION-RETRIEVAL

This paper describes a development of a spoken dialogue travel guidanc e system, TARSAN. TARSAN uses commercial CD-ROM guidebooks as its know ledge source, containing a large amount of travel information. To deal with this amount of information, a large vocabulary has to be accepte d by a speech recognizer without reducing its performance. Thus, we pr opose two steps of active/non-active word control methods: (1) a word/ grammar prediction strategy, and (2) unknown word re-evaluation algori thm. The word/grammar prediction strategy dynamically changes a recogn ition network according to a conversation situation by making use of r esults retrieved from the CD-ROMs. This strategy makes users to access almost all data on the CD-ROMs using a small vocabular speech recogni zer. The unknown word re-evaluation algorithm processes unknown words and non-active words using Garbage Models by integrating them into the recognition network, and once the Garbage Models are recognized, the unknown part will be compared with the non-active words. This algorith m enhances the ability of the word/grammar prediction. In the experime nt without Garbage Models, 80.9% of the utterances were correctly unde rstood. In the unknown word re-evaluation experiment using the Garbage Models, 86.4% were correctly re-evaluated, while the false alarms of 5% were found.