Abstract | ||
---|---|---|
Previous work has demonstrated the success of statistical lan- guage models when enough training data is available (1), but despite that, grammar-based systems are proving the preferred choice in successful commercial systems such as HeyAnita (2), BeVocal (3) and Tellme (4), largely due to the difficulty in- volved in obtaining a corpus of training data. Here we trained an SLM on data obtained using a grammar-based system and compared the performance of the two systems with regards to recognition. We also parsed the output of the SLM using a ro- bust parser and compared the accuracy of the semantic output of the systems. The SLM/robust parser showed considerable im- provement on unconstrained input, and similar precision/recall (per slot value) on utterances provided by trained users. |
Year | Venue | Field |
---|---|---|
2001 | INTERSPEECH | Training set,Computer science,Speech recognition,Grammar,Natural language processing,Artificial intelligence,Parsing,Recall,Language model |
DocType | Citations | PageRank |
Conference | 31 | 3.18 |
References | Authors | |
4 | 6 |
Name | Order | Citations | PageRank |
---|---|---|---|
Sylvia Knight | 1 | 31 | 3.18 |
Genevieve Gorrell | 2 | 266 | 22.00 |
Manny Rayner | 3 | 508 | 89.27 |
David Milward | 4 | 196 | 27.51 |
Rob Koeling | 5 | 434 | 38.38 |
Ian Lewin | 6 | 246 | 25.58 |