As of December 2006 I have been working as a P.h.D student at Radboud University Nijmegen. I am enrolled in the MIDAS project, focussing on noise robust automatic speech recognition (ASR) using missing data masking.
My research revolves around doing ASR in situations where the speech signal is distorted by background noise. Noise reduction techniques have been applied with some success in the past, but there remains a large performance gap between the best ASR implementations and human recognition, especially when the noise is non-stationary. This project tackles the noise robustness problem in ASR through missing data techniques (MDT). In this approach, every time-frequency cell is estimated to be either reliable (clean) or unreliable (noisy). The unreliable parts are then imputed (estimated) before performing recognition.
The institute I work at can be found here. My daily supervisors are Bert Cranen and Louis ten Bosch
Other research projects
I am currently working on Compressive Sensing techniques for use in classification and data imputation in speech.
Research Interests
Speech recognition, Robustness, Machine learning, Compressive Sensing, Python Programming