Performance of speech recognition systems strongly degrades in the presence of background noise, like the driving noise in the interior of a car. The authors compare two different Kalman filtering approaches which attempt to improve noise robustness: Switching Linear Dynamic Models (SLDM) and Autoregressive Switching Linear Dynamical Systems (ARSLDS). Unlike previous works which are restricted on considering white noise, the authors evaluate the modeling concepts in a noisy speech recognition task where also colored noise produced through different driving conditions and car types is taken into account. Thereby they demonstrate that speech enhancement based on Kalman filtering prevails over all standard de-noising techniques considered herein, such as Wiener filtering, Histogram Equalization, and Unsupervised Spectral Subtraction.
Switching linear dynamic models for noise robust in-car speech recognition
2008
10 Seiten, 7 Bilder, 1 Tabelle, 20 Quellen
Aufsatz (Konferenz)
Englisch
Extended VTS for noise-robust speech recognition
TIBKAT | 2009
|Noise Robust Speech Recognition Using Prosodic Information
Springer Verlag | 2005
|Noise robust speech recognition and coding in car environment
Kraftfahrwesen | 1991
|Noise Robust Speech Recognition Makes In-Car Navigation Safe and Affordable
British Library Conference Proceedings | 1995
|Noise robust speech recognition makes in-car navigation safe and affordable
Tema Archiv | 1995
|