Output details
11 - Computer Science and Informatics
University of Sheffield
Article title
Combining Speech Fragment Decoding and Adaptive Noise Floor Modeling
Type
D - Journal article
Title of journal
IEEE Transactions on Audio, Speech, and Language Processing
Article number
-
Volume number
20
Issue number
3
First page of article
818
ISSN of journal
15587924
Year of publication
2012
URL
-
Number of additional authors
3
Additional information
<22> The paper presents a perceptually-inspired approach to robust speech recognition drawing upon research conducted at Sheffield over a 15 year period. The developments reported here result from a recently completed EPSRC project (CHIME), rated as `outstanding’ in its final review. The paper is the first to show how a model of perceptual processing can be employed to deliver state-of-the-art speech recognition results in a challenging distant-microphone evaluation. The model has been adopted by partners of the Marie Curie ITN, INSPIRE as the basis of a model of speech intelligibility.
Interdisciplinary
-
Cross-referral requested
-
Research group
None
Citation count
1
Proposed double-weighted
No
Double-weighted statement
-
Reserve for a double-weighted output
No
Non-English
No
English abstract
-
