Research
My research interests are directed towards statistical signal processing, spatial acoustics, machine learning, and their application in speech communication. In particular, my research areas of interest are:
Speech Enhancement
Speaker Separation
Speaker Identification
Acoustic Echo Control
Noise Cancellation
Acoustic Beamforming
Estimation Theory
Machine Learning
MIMO Signal Processing
Development
Many of my research interests are motivated by industrial applications in the fields of digital signal processing, VoIP systems, and control systems. In these areas, rapid prototyping is the key towards fast time-to-market. My approach encompasses three crucial competencies:
Algorithm Design:
Disruptive technologies always start with extensive research. Hence it is essential to translate research results into simulation code using Matlab, Python, and Hardware-In-the-Loop (HIL) Verification and Validation tools.
Software Design:
To run efficiently, each algorithm needs to be optimized to its targeted platform. For example, optimizations can be done by using hardware acceleration such as ARM/NEON, or using dedicated frameworks such as TFLite.
Hardware Design:
The hardware platform lives at the intersection of the demands of the algorithm, and the economic frame of the application. It is therefore crucial to tailor an embedded system to the specific requirements of each project.