Audiovisual Speech Source Separation
- Submitting institution
-
University of Newcastle upon Tyne
- Unit of assessment
- 12 - Engineering
- Output identifier
- 211478-197546-1293
- Type
- D - Journal article
- DOI
-
10.1109/MSP.2013.2296173
- Title of journal
- IEEE Signal Processing Magazine
- Article number
- -
- First page
- 125
- Volume
- 31
- Issue
- 3
- ISSN
- 1053-5888
- Open access status
- Compliant
- Month of publication
- April
- Year of publication
- 2014
- URL
-
http://dx.doi.org/10.1109/MSP.2013.2296173
- 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
- No
- Number of additional authors
-
3
- Research group(s)
-
-
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- A framework for overdetermined, determined, and underdetermined speech processing of multiple concurrent speakers, which has direct applications in healthcare and human machine interface. The key challenge was in dealing with simultaneous, physically stationary and moving, speakers in highly reverberant room environments. Joint audio-visual data processing techniques based on statistical signal processing and computational auditory scene analysis. Outcome of international collaboration with University of Grenoble France and the EPSRC project (EP/H049665/1:£660K), led to the dstl/EPSRC University Defence Research Collaboration in Signal Processing award (EP/K014307:£3.4M), also led to EPSRC grant (EP/V002856/1:£865K).
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -