Abstract
In spoken dialog systems it is often the case that the sentence produced by the decoder with the highest recognition probability may not be the best choice for extracting the intended concepts. Lower ranking hypotheses may present better alternatives. In this paper, we show how to integrate multiple knowledge sources for the decision of selecting one of these hypotheses. A scoring schema combining information from the recognizer output, the parser, an utterance type classifier and dialog context is used. The scaling weights of the combined scores are determined automatically by an optimization procedure. Finally, we show the results of testing this approach and its performance compared to the approach of selecting the best recognition hypothesis.
| Original language | English (US) |
|---|---|
| Title of host publication | EUROSPEECH 2001 - SCANDINAVIA - 7th European Conference on Speech Communication and Technology |
| Publisher | International Speech Communication Association |
| Pages | 1783-1786 |
| Number of pages | 4 |
| ISBN (Electronic) | 8790834100, 9788790834104 |
| State | Published - 2001 |
| Event | 7th European Conference on Speech Communication and Technology - Scandinavia, EUROSPEECH 2001 - Aalborg, Denmark Duration: Sep 3 2001 → Sep 7 2001 |
Other
| Other | 7th European Conference on Speech Communication and Technology - Scandinavia, EUROSPEECH 2001 |
|---|---|
| Country | Denmark |
| City | Aalborg |
| Period | 9/3/01 → 9/7/01 |
Fingerprint
ASJC Scopus subject areas
- Communication
- Linguistics and Language
- Computer Science Applications
- Software
Cite this
Integrating multiple knowledge sources for improved speech understanding. / Abdou, Sherif; Scordilis, Michael S.
EUROSPEECH 2001 - SCANDINAVIA - 7th European Conference on Speech Communication and Technology. International Speech Communication Association, 2001. p. 1783-1786.Research output: Chapter in Book/Report/Conference proceeding › Conference contribution
}
TY - GEN
T1 - Integrating multiple knowledge sources for improved speech understanding
AU - Abdou, Sherif
AU - Scordilis, Michael S
PY - 2001
Y1 - 2001
N2 - In spoken dialog systems it is often the case that the sentence produced by the decoder with the highest recognition probability may not be the best choice for extracting the intended concepts. Lower ranking hypotheses may present better alternatives. In this paper, we show how to integrate multiple knowledge sources for the decision of selecting one of these hypotheses. A scoring schema combining information from the recognizer output, the parser, an utterance type classifier and dialog context is used. The scaling weights of the combined scores are determined automatically by an optimization procedure. Finally, we show the results of testing this approach and its performance compared to the approach of selecting the best recognition hypothesis.
AB - In spoken dialog systems it is often the case that the sentence produced by the decoder with the highest recognition probability may not be the best choice for extracting the intended concepts. Lower ranking hypotheses may present better alternatives. In this paper, we show how to integrate multiple knowledge sources for the decision of selecting one of these hypotheses. A scoring schema combining information from the recognizer output, the parser, an utterance type classifier and dialog context is used. The scaling weights of the combined scores are determined automatically by an optimization procedure. Finally, we show the results of testing this approach and its performance compared to the approach of selecting the best recognition hypothesis.
UR - http://www.scopus.com/inward/record.url?scp=32144441480&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=32144441480&partnerID=8YFLogxK
M3 - Conference contribution
SP - 1783
EP - 1786
BT - EUROSPEECH 2001 - SCANDINAVIA - 7th European Conference on Speech Communication and Technology
PB - International Speech Communication Association
ER -