An Instrumental Intelligibility Metric Based on Information Theory

Research output: Contribution to journalArticleScientificpeer-review

38 Citations (Scopus)

Abstract

We propose a monaural intrusive instrumental intelligibility metric called SIIB (speech intelligibility in bits). SIIB is an estimate of the amount of information shared between a talker and a listener in bits per second. Unlike existing information theoretic intelligibility metrics, SIIB accounts for talker variability and statistical dependencies between time-frequency units. Our evaluation shows that relative to state-of-the-art intelligibility metrics, SIIB is highly correlated with the intelligibility of speech that has been degraded by noise and processed by speech enhancement algorithms.

Original languageEnglish
Pages (from-to)115-119
Number of pages5
JournalIEEE Signal Processing Letters
Volume25
Issue number1
DOIs
Publication statusPublished - 2018

Keywords

  • Intelligibility
  • mutual information

Fingerprint

Dive into the research topics of 'An Instrumental Intelligibility Metric Based on Information Theory'. Together they form a unique fingerprint.

Cite this