Fraunhofer-Gesellschaft

Publica

Hier finden Sie wissenschaftliche Publikationen aus den Fraunhofer-Instituten.

Employing stochastic constrained LMS algorithm for ASR frontend processing

 
: Stadtschnitzer, Michael; Stein, Daniel; Bardeli, Rolf

:
Volltext ()

The 2nd CHiME Workshop on Machine Listening in Multisource Environments 2013. USB-Stick : 1st June 2013, Vancouver, Canada. In conjunction with ICASSP 2013
Vancouver, 2013
S.71-72
International Workshop on Machine Listening in Multisource Environments <2, 2013, Vancouver>
International Conference on Acoustics, Speech, and Signal Processing (ICASSP) <38, 2013, Vancouver>
Englisch
Konferenzbeitrag, Elektronische Publikation
Fraunhofer IAIS ()

Abstract
In scenarios with multiple input single output systems, the stochastic constrained least mean-squares (LMS) algorithmhas been proven to be an effective approach. However, whenonly two input channels are available, it is unclear whetherthis approach still yields improvements. In this paper, we investigate the stableness and the robustness of the constrained LMS algorithm on âTrack 1â of â2nd CHiME Challengeâ and show that it leads to small yet consistent improvements on all signal-to-noise settings.

: http://publica.fraunhofer.de/dokumente/N-415194.html