Noise Power Spectral Density Estimation Using MaxNSR Blocking Matrix
- Submitting institution
-
Queen Mary University of London
- Unit of assessment
- 11 - Computer Science and Informatics
- Output identifier
- 408
- Type
- D - Journal article
- DOI
-
10.1109/TASLP.2015.2438542
- Title of journal
- IEEE Transactions on Audio, Speech and Language Processing
- Article number
- -
- First page
- 1493
- Volume
- 23
- Issue
- 9
- ISSN
- 1558-7916
- Open access status
- Out of scope for open access requirements
- Month of publication
- June
- Year of publication
- 2015
- 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
- No
- Number of additional authors
-
2
- Research group(s)
-
-
- Citation count
- 23
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- The work proposed a post-filtering technique to reduce the residual noise after microphone-array spatial filtering. This work theoretically derived the relationship between the spatial filter and the post-filter, and thus outperforms the state-of-the-art method, which does not consider this relationship. The work lead to submissions to the Signal Separation Evaluation Campaign (SISEC2013 (https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=6661988) and achieved top results out of 6 entries in the task of real-world background noise reduction. The work was cited by a patent (https://patents.google.com/patent/US9607603B1/en). This work was supported by the world-recognized Alexander von Humboldt Fellowship (?100,000, 24 months) from the Humboldt Foundation Germany (https://www.humboldt-foundation.de).
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -