Research group:
- Speech and Dialog Research Group at MSR (2016-)
Research interests:
- Audio source separation
- Far-field speech recognition
Labs affiliated with at Sabanci University (2002-2016):
Past projects:
- “Speaker and noise specific speech denoising for telecommunication systems,” supported by Turk Telekom, Dates: Mar. 2013-Mar. 2014, Role: PI.
- “Biometric data collection and system improvement ,” supported by TUBITAK BILGEM, Dates: Sep 2012 – Sep 2013, Role: PI.
- “Single channel source separation,” supported by Turk Telekom, Dates: Mar. 2012-Mar. 2013, Budget: 66000 TL, Role: PI.
- “Fast and convergent algorithms for large scale learning (1001-110E041)”, supported by TUBITAK, Dates: Oct. 2010 – Sep. 2012, Budget: 170000 TL, Role: PI.
- “Novel approaches in audio-visual speech recognition,” supported by TUBITAK, Dates: Oct. 2007 – Sep. 2010, Budget: 221945 YTL, Role: PI.
- “Speaker recognition using support vector machines,” supported by TUBITAK UEKAE, Dates: May 2008 – May 2009, Role: PI.
- “Driving behavior analysis,” supported by NEDO, Japan, Dates: 2006-2009, Budget (for Sabanci only): 11.7 million JPY (100K USD), Role: PI for Sabanci University part of the project since Winter 2007.
- “Drive-safe,” supported by DPT, Turkey, Dates: 2006-2009, Budget (total) approximately 1800000 YTL, Role: researcher.
Past research Interests:
- Pattern recognition and machine learning especially applied to multimedia and audio signals
- Speech processing (separation, enhancement)
- Speech and speaker recognition
- Compressed sensing signal recovery
- Optimization algorithms for machine learning, image reconstruction in tomography and other applications
- Biometrics involving voice and face identification/verification