Title | ||
---|---|---|
Robust Parsing of Spoken Dialogue Using Contextual Knowledge and Recognition Probabilities |
Abstract | ||
---|---|---|
In this paper we describe the linguistic processingcomponent of a spoken dialogue system. The task ofthis word graph parser is to find the most plausiblesequence of word hypotheses in the input graph. Ifno global solution can be found, a robust mechanismof selecting multiple partial results is applied. We arguethat the semantic accuracy of the selected resultscan be considerably improved if the selection is basedon an integrated quality score combining word recognitionprobabilities and ... |
Year | Field | DocType |
---|---|---|
1995 | Graph,Quality Score,Computer science,Word recognition,Speech recognition,Artificial intelligence,Natural language processing,Parsing,Word graph | Journal |
Volume | Citations | PageRank |
cmp-lg/9505017 | 4 | 0.60 |
References | Authors | |
0 | 2 |
Name | Order | Citations | PageRank |
---|---|---|---|
Gunther Gorz | 1 | 5 | 0.99 |
Gerhard Hanrieder Bavarian | 2 | 4 | 0.60 |