Robust detection of voiced segments in samples of everyday conversations using unsupervised HMMS

Meysam Asgari, Izhak Shafran, Alireza Bayestehtashk

Research output: Chapter in Book/Report/Conference proceedingConference contribution

6 Scopus citations

Abstract

We investigate methods for detecting voiced segments in everyday conversations from ambient recordings. Such recordings contain high diversity of background noise, making it difficult or infeasible to collect representative labelled samples for estimating noise-specific HMM models. The popular utility get-f0 and its derivatives compute normalized cross-correlation for detecting voiced segments, which unfortunately is sensitive to different types of noise. Exploiting the fact that voiced speech is not just periodic but also rich in harmonic, we model voiced segments by adopting harmonic models, which have recently gained considerable attention. In previous work, the parameters of the model were estimated independently for each frame using maximum likelihood criterion. However, since the distribution of harmonic coefficients depend on articulators of speakers, we estimate the model parameters more robustly using a maximum a posteriori criterion. We use the likelihood of voicing, computed from the harmonic model, as an observation probability of an HMM and detect speech using this unsupervised HMM. The one caveat of the harmonic model is that they fail to distinguish speech from other stationary harmonic noise. We rectify this weakness by taking advantage of the non-stationary property of speech. We evaluate our models empirically on a task of detecting speech on a large corpora of everyday speech and demonstrate that these models perform significantly better than standard voice detection algorithm employed in popular tools.

Original languageEnglish (US)
Title of host publication2012 IEEE Workshop on Spoken Language Technology, SLT 2012 - Proceedings
Pages438-442
Number of pages5
DOIs
StatePublished - 2012
Event2012 IEEE Workshop on Spoken Language Technology, SLT 2012 - Miami, FL, United States
Duration: Dec 2 2012Dec 5 2012

Publication series

Name2012 IEEE Workshop on Spoken Language Technology, SLT 2012 - Proceedings

Conference

Conference2012 IEEE Workshop on Spoken Language Technology, SLT 2012
Country/TerritoryUnited States
CityMiami, FL
Period12/2/1212/5/12

Keywords

  • life log
  • speech detection
  • voice detection

ASJC Scopus subject areas

  • Language and Linguistics
  • Linguistics and Language

Fingerprint

Dive into the research topics of 'Robust detection of voiced segments in samples of everyday conversations using unsupervised HMMS'. Together they form a unique fingerprint.

Cite this