Speech and Vision Lab

  • Increase font size
  • Default font size
  • Decrease font size
Home Publications
A comparative study of constrained and unconstrained approaches for segmentation of speech signal
Research Area: Speech Analysis Year: 2010
Type of Publication: In Proceedings Keywords: HMM, GroupDelay, ANN, Speech Segmentation
Authors: Venkatesh Keri, Kishore S. Prahallad  
In this work, we compare different approaches for speech segmentation, of which some are constrained and the remaining are unconstrained by phone transcript. A high accuracy speech segmentation can be obtained by approaches constrained by phone transcript such as HMM forced-alignment when exact phone transcript is known. But such approaches have to adjust with canonical phone transcript, as exact phone transcript is tough to obtain. Our experiments on TIMIT corpus demonstrate that ANN and HMM phone-loop based unconstrained approaches, perform better than HMM forced-alignment based approach constrained by canonical phone transcript. Finally a detailed error analysis of these approaches is reported.
Digital version