Paper: Word Fragments Identification Using Acoustic-Prosodic Features In Conversational Speech

ACL ID N03-3007
Title Word Fragments Identification Using Acoustic-Prosodic Features In Conversational Speech
Venue Human Language Technologies
Session Student Session
Year 2003
Authors
  • Yang Liu (University of California at Berkeley, Berkeley CA; Purdue University, West Lafayette IN)

Word fragments pose serious problems for speech recognizers. Accurate identification of word fragments will not only improve recogni- tion accuracy, but also be very helpful for dis- fluency detection algorithm because the occur- rence of word fragments is a good indicator of speech disfluencies. Different from the previ- ous effort of including word fragments in the acoustic model, in this paper, we investigate the problem of word fragment identification from another approach, i.e. building classifiers using acoustic-prosodic features. Our experiments show that, by combining a few voice quality measures and prosodic features extracted from the forced alignments with the human tran- scriptions, we obtain a precision rate of 74.3% and a recall rate of 70.1% on the downsampled data of sponta...