CochleaNet: A robust language-independent audio-visual model for real-time speech enhancement
- Submitting institution
-
Edinburgh Napier University
- Unit of assessment
- 11 - Computer Science and Informatics
- Output identifier
- 2692707
- Type
- D - Journal article
- DOI
-
10.1016/j.inffus.2020.04.001
- Title of journal
- Information Fusion
- Article number
- -
- First page
- 273
- Volume
- 63
- Issue
- -
- ISSN
- 1566-2535
- Open access status
- Compliant
- Month of publication
- April
- Year of publication
- 2020
- URL
-
-
- Supplementary information
-
-
- Request cross-referral to
- -
- Output has been delayed by COVID-19
- No
- COVID-19 affected output statement
- -
- Forensic science
- No
- Criminology
- No
- Interdisciplinary
- Yes
- Number of additional authors
-
3
- Research group(s)
-
-
- Citation count
- 1
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- This pioneering paper represents a body of work from the internationally acclaimed EPSRC grant (AV-COGHEAR:EP/M026981/1, widely covered in the media, e.g. http://www.bbc.co.uk/news/uk-scotland-tayside-central-33098322), introducing the world’s first language-independent speech-enhancement model for lip-reading hearing-aids. A first real-world benchmark audio-visual (AV) corpus is also developed (ASPIRE:https://zenodo.org/record/4585619#.YFp4i0j7Su4), stimulating AV speech-processing research and innovation. Our model underpins an EPSRC transformative healthcare technologies programme grant (COG-MHEAR:EP/T021063/1). This has brought together seven Universities, and a User-Group (including hearing-aid manufacturers, Sonova and Nokia-Bell Labs), committing ~£850k matched funding, to develop and commercialise a real-time demonstrator, enabled through integration of our AV model with 5G, IoT, cybersecurity and skin-electronics.
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -