For the current REF see the REF 2021 website REF 2021 logo

Output details

13 - Electrical and Electronic Engineering, Metallurgy and Materials

University of Birmingham : A - Electronic, Electrical and computer engineering

Return to search Previous output Next output
Output 47 of 108 in the submission
Article title

Human and computer recognition of regional accents and ethnic groups from British English speech

Type
D - Journal article
Title of journal
Computer Speech and Language
Article number
-
Volume number
27
Issue number
1
First page of article
59
ISSN of journal
0885-2308
Year of publication
2013
URL
-
Number of additional authors
2
Additional information

This paper is concerned with automatic identification of regional accents, and ethnic groups within an accent, from speech. The method is calibrated on the international US-NIST-2003 Language Identification standard task and is state-of-the-art. Accent recognition performance is superior to any published result, and recognition of ethnic groups has not been attempted previously. Crucial speech data was collected with government funding by our spin-out company, The Speech Ark. Expertise in detection technology was key in securing £623k long-term funding from GCHQ, and PhD funding from Novauris Limited. The same techniques are applied to stroke rehabilitation in the EU CogWatch project.

Interdisciplinary
-
Cross-referral requested
-
Research group
B - Human Computer Interaction
Proposed double-weighted
No
Double-weighted statement
-
Reserve for a double-weighted output
No
Non-English
No
English abstract
-