Output details
11 - Computer Science and Informatics
King's College London
Return to search
Output 0 of 0 in the submission
Article title
Combined Features and Kernel Design for Noise Robust Phoneme Classification Using Support Vector Machines
Type
D - Journal article
Title of journal
Ieee Transactions On Audio Speech And Language Processing
Article number
5618550
Volume number
19
Issue number
5
First page of article
1396
ISSN of journal
1558-7916
Year of publication
2011
URL
-
Number of additional authors
3
Additional information
<22>The paper presents a paradigm shift in automatic speech recognition (ASR). ASR systems suffer from a dramatic performance degradation in the presence of additive noise or a mismatch between training and testing conditions. A paradigm of representing speech in high-dimensional spaces of acoustic waveforms of speech is introduced and developed in the paper. It is further demonstrated that it enables significant improvements in robustness of ASR. This represents a groundbreaking insight considering that all ASR systems operate in low-dimensional feature spaces.
Interdisciplinary
-
Cross-referral requested
-
Research group
F - Centre for Telecommunications Research
Citation count
1
Proposed double-weighted
No
Double-weighted statement
-
Reserve for a double-weighted output
No
Non-English
No
English abstract
-
