Paper: Class Model Adaptation For Speech Summarisation

ACL ID N06-2006
Title Class Model Adaptation For Speech Summarisation
Venue Human Language Technologies
Session Short Paper
Year 2006
Authors

The performance of automatic speech summarisation has been improved in pre- vious experiments by using linguistic model adaptation. We extend such adapta- tion to the use of class models, whose ro- bustness further improves summarisation performance on a wider variety of objec- tive evaluation metrics such as ROUGE-2 and ROUGE-SU4 used in the text sum- marisation literature. Summaries made from automatic speech recogniser tran- scriptions benefit from relative improve- ments ranging from 6.0% to 22.2% on all investigated metrics.