Sebastian Gergen

Learn More
Audio signal classification suffers from the mismatch of environmental conditions when training data is based on clean and anechoic signals and test data is distorted by reverberation and signals from other sources. In this contribution we analyze the classification performance for such a scenario with two concurrently active sources in a simulated(More)
Automatic speech recognition (ASR) enables very intuitive human-machine interaction. However, signal degradations due to reverberation or noise reduce the accuracy of audio-based recognition. The introduction of a second signal stream that is not affected by degradations in the audio domain (e.g., a video stream) increases the robustness of ASR against(More)
In 2011, Borß introduced a parametric model for the design of virtual acoustics, which creates a natural sounding virtual environment for applications requiring virtualization e.g., in teleconferencing systems and computer games. In this work we refine this model to make it applicable for the simulation of room acoustics and reverberation to aid in(More)
The classification of acoustic signals is an important step in many audio signal processing algorithms, e.g. in the context of speech enhancement, speech recognition, and others. Signals which are captured for classification are often degraded by an unknown amount of reverberation in a real environment. If a classifier is trained on clean and anechoic data,(More)
We present here a hierarchical approach for the detection and localisation of brake squeal. The proposed system exploits the spatial diversity of microphone arrays to localise a squealing brake. As brake squeal is emitted from a priori known regions, i.e. near the wheels, localisation of a squeal may be seen as a hypothesis testing problem. However, in(More)