Combining Monaural Beamforming and Blind Source Separation for Binaural Speech Enhancement in Multi-Microphone Hearing Aids
Konferenz: Sprachkommunikation 2010 - 9. ITG-Fachtagung
06.10.2010 - 08.10.2010 in Bochum, Deutschland
Tagungsband: Sprachkommunikation 2010
Seiten: 4Sprache: EnglischTyp: PDF
Persönliche VDE-Mitglieder erhalten auf diesen Artikel 10% Rabatt
Autoren:
Reindl, Klaus; Zheng, Yuanhang; Kellermann, Walter (University of Erlangen-Nuremberg, Multimedia Communications and Signal Processing, Cauerstr. 7, 91058 Erlangen, Germany)
Prokein, Peter; Fischer, Eghart (Siemens Audiologische Technik GmbH, Research & Development Signal Processing (RD SP), Gebbertstr. 125, 91058 Erlangen, Germany)
Inhalt:
A binaural speech enhancement concept based on blind source separation (BSS) formulti-microphone hearing aids is introduced. The proposed concept combines monaural differential beamformers and a blind binaural speech enhancement strategy for improved noise suppression. In contrast to a BSS-based speech enhancement strategy with only two microphones, this scheme solves the front-back ambiguity of two-microphone binaural systems. This novel binaural speech enhancement concept is analyzed and evaluated for different scenarios. The proposed combination turns out to be an efficient concept for binaural suppression of interfering point sources as well as background noise.