My research is focused on speech recognition and language understanding, in which I develop brain-inspired models based on echo state networks to be able to explain functional principles in the human brain and exploit their benefits. In the last seven years, I was working on a system combining speech recognition and semantics which produces hypotheses in a cooperative way. I am also supervising the DOCKS project, which is focused on post-processing hypotheses from Google's cloud-based speech recognition system. Currently, I am working for the EXIST project LemonSpeech funded by the Bundesministerium für Wirtschaft und Energie. It aims to transfer the research from English automatic speech recognition (ASR) to German ASR and evaluate the possibility for a market entry. Also, further research is performed to reach state-of-the-art performance in German ASR. The novel models aim to achieve equivalent performance as computational intense Cloud-based ASR models, while being runnable on inexpensive local hardware. The models developed in the last years are evidently runnable on this hardware, future models will be adapted to it.
Interests:
- Reservoir Computing and Echo State networks
- Brain-Inspired Systems, Machine Learning and Neural Networks in general
- Speech Recognition and Language Understanding
- Deep Learning
