Ernst Seidel

Ernst Seidel serves as a Researcher in the Department of Signal Processing and Machine Learning at the Institute of Communications Engineering, Technical University of Braunschweig. Since January 2021, he has contributed to cutting-edge research in speech and audio signal processing within the Faculty of Electrical Engineering, Information Technology, and Physics.

His academic foundation includes:

• Master's in Electrical Engineering (Communications Technology specialization), TU Braunschweig (2018-2020)
• Bachelor's in Electrical Engineering (Communications Technology specialization), TU Braunschweig (2014-2018)
• General higher education entrance qualification, GutsMuths-Gymnasium Quedlinburg (2014)

Seidel's research focuses on machine learning-driven signal processing for real-time speech applications. His work pioneers low-latency deep adaptive filters for hearing aid systems and DNN-based acoustic echo cancellation using Kalman filter postprocessing. He investigates user-controllable echo suppression systems that balance near-end speech quality with echo reduction, addressing critical challenges in stereo channel processing and real-world deployment constraints.

As a thesis supervisor, Seidel guides master's students in developing neural network architectures for acoustic echo cancellation, emphasizing practical implementation requirements including Python/MATLAB proficiency and foundational knowledge in digital signal processing. His supervision reflects the department's industry-aligned research priorities in real-time audio systems.

Seidel operates within the Department of Signal Processing and Machine Learning's research ecosystem, collaborating on projects requiring stringent latency constraints and computational efficiency. His work bridges theoretical machine learning advances with commercial audio applications through partnerships with industry leaders like Cerence and Far Field Sound Corporation.