Achtung:

Sie haben Javascript deaktiviert!
Sie haben versucht eine Funktion zu nutzen, die nur mit Javascript möglich ist. Um sämtliche Funktionalitäten unserer Internetseite zu nutzen, aktivieren Sie bitte Javascript in Ihrem Browser.

Show image information
Tuesday, 29.01.2019 | 16.00 Uhr - 18.00 Uhr | Location: P6.2.03

Integration of neural networks and probabilistic models for acoustic blind source separation

Abstract

Current trends and the astonishing performance of neural networks can hardly be neglected. Besides their long lasting success in speech recognition many deep learning solutions for speech enhancement and more recently also for speech separation are available. However, when a geometrical or physical interpretation of a signal is possible statistical models tend to be more interpretable and yield reasonably good solutions in many cases. This presentation demonstrates how neural networks and probabilistic models can be integrated or profit from each other for the specific example of multi channel acoustic blind source separation.

Speaker

Lukas Drude

The University for the Information Society