Main area of research
Felix Weninger is a principal research scientist in acoustic modeling at Nuance Communications in Ulm, Germany. His research interests include deep learning, speech recognition, and source separation. He received his PhD degree in computer science (Dr. rer. nat.) from Technische Universität München (TUM), one of the German Excellence Universities, in 2015. From 2010 through 2014, he worked as a research assistant in the Machine Intelligence and Signal Processing Group at TUM's Institute for Human-Machine Communication, focusing on new machine learning approaches for noise-robust automatic speech recognition and related tasks. In 2013/14, he interned at Mitsubishi Electric Research Labs (MERL), Cambridge, MA, USA, where he developed deep learning based speech enhancement. Dr. Weninger has published more than 90 peer-reviewed papers in books, journals and conference proceedings, which can be found here. He also has a Google Scholar profile.
We evaluate some recent developments in recurrent neural network (RNN) based speech enhancement in the light of noise-robust automatic speech recognition (ASR). The proposed framework …