Loading...
Please wait, while we are loading the content...
Similar Documents
Discriminative feature weighting using MCE training for topic identification of spoken audio recordings (2008)
| Content Provider | CiteSeerX |
|---|---|
| Author | Margolis, Anna Hazen, Timothy J. |
| Description | in Proc. IEEE ICASSP, Las Vegas, NV |
| Abstract | In this paper we investigate a discriminative approach to feature weighting for topic identification using minimum classification error (MCE) training. Our approach learns feature weights by optimizing an objective loss function directly related to the classification error rate of the topic identification system. Topic identification experiments are performed on spoken conversations from the Fisher corpus. Features drawn from both word and phone lattices generated via automatic speech recognition are investigated. Under various different conditions, our new feature weighting scheme reduces our classification error rate between 9 % and 23 % relative to our baseline naive Bayes system using feature selection. Index Terms — Audio document processing, topic identification, topic spotting. 1. |
| File Format | |
| Publisher Date | 2008-01-01 |
| Access Restriction | Open |
| Subject Keyword | Automatic Speech Recognition Discriminative Approach Minimum Classification Error Feature Weight Topic Identification Experiment Index Term Audio Document Processing Various Different Condition Baseline Naive Bayes System Topic Identification Classification Error Rate New Feature Topic Identification System Topic Spotting Objective Loss Function Feature Selection Fisher Corpus Spoken Conversation Phone Lattice Spoken Audio Recording Mce Training Discriminative Feature |
| Content Type | Text |
| Resource Type | Article |
