Semi-non-intrusive objective intelligibility measure using spatial filtering in hearing aids

Charlotte Sørensen, Jesper Bünsow Boldt, Frederik Gran, Mads Græsbøll Christensen

Research output: Contribution to book/anthology/report/conference proceedingArticle in proceedingResearchpeer-review

9 Citations (Scopus)
324 Downloads (Pure)

Abstract

Reliable non-intrusive online assessment of speech intelligibility can play a key role for the functioning of hearing aids, e.g. as guidance for adjusting the hearing aid settings to the environment. While existing intrusive metrics can provide a precise and reliable measure, the current non-intrusive metrics have not been able to achieve acceptable intelligibility predictions. This paper presents a new semi-non-intrusive intelligibility measure based on an existing intrusive measure, STOI, where an estimate of the clean speech is extracted using spatial filtering in the hearing aid. The results indicate that the STOI score obtained with the proposed method using an estimate of the clean speech correlates well with the STOI score having the original clean speech signal available.
Original languageEnglish
Title of host publication24th European Signal Processing Conference (EUSIPCO), 29 August - 2 September 2016, Budapest, Hungary
PublisherIEEE
Publication date2016
Pages1358-1362
ISBN (Print)978-0-9928-6265-7
DOIs
Publication statusPublished - 2016
Event24th European Signal Processing Conference - Hilton Budapest, Budapest, Hungary
Duration: 29 Aug 20162 Sept 2016
Conference number: 24

Conference

Conference24th European Signal Processing Conference
Number24
LocationHilton Budapest
Country/TerritoryHungary
CityBudapest
Period29/08/201602/09/2016
SeriesProceedings of the European Signal Processing Conference
ISSN2076-1465

Keywords

  • Non-intrusive intellgibility measure
  • generalized sidelobe canceller
  • hearing aids

Fingerprint

Dive into the research topics of 'Semi-non-intrusive objective intelligibility measure using spatial filtering in hearing aids'. Together they form a unique fingerprint.

Cite this