Paper: Active Learning For Classifying Phone Sequences From Unsupervised Phonotactic Models

ACL ID N03-2007
Title Active Learning For Classifying Phone Sequences From Unsupervised Phonotactic Models
Venue Human Language Technologies
Session Short Paper
Year 2003
Authors

This paper describes an application of active learning methods to the classification of phone strings recognized using unsupervised phono- tactic models. The only training data required for classification using these recognition meth- ods is assigning class labels to the audio files. The work described here demonstrates that substantial savings in this effort can be ob- tained by actively selecting examples to be la- beled using confidence scores from the Boos- Texter classifier. The saving in class label- ing effort is evaluated on two different spo- ken language system domains in terms both of the number of utterances to be labeled and the length of the labeled utterances in phones. We show that savings in labeling effort of around 30% can be obtained using active selection of examples.