You are in : Home » Results & submissions » Select UOA » 11 - Computer Science and Informatics » View submission: University of Sheffield » Outputs » Detail

Output details

11 - Computer Science and Informatics

University of Sheffield

Return to search Previous output Next output

Output 41 of 109 in the submission

Article title

Energetic and Informational Masking Effects in an Audiovisual Speech Recognition System

Type

D - Journal article

DOI

10.1109/TASL.2008.2011534

Title of journal

IEEE Transactions on Audio, Speech, and Language Processing

Article number

Volume number

Issue number

First page of article

446

ISSN of journal

15587916

Year of publication

2009

URL

Number of additional authors

Additional information

<22> The paper fulfils the key objective of an EPSRC first-grant scheme project (GR/T04823/01) - a project that was given an overall assessment of 'outstanding' in its final review. The paper is the first to demonstrate via a computational model how visual speech signals (e.g. lip movements) may increase speech intelligibly, not only by providing complementary linguistic information, but also by affording improved audio source separation. The work has implications for hearing-aid which are being discussed with industrial partners within the framework of the Marie-Curie ITN INSPIRE.

Interdisciplinary

Cross-referral requested

Research group

None

Citation count

Proposed double-weighted

Double-weighted statement

Reserve for a double-weighted output

Non-English

English abstract