Non-stationary feature extraction for automatic speech recognition

In current speech recognition systems mainly Short-Time Fourier Transform based features like MFCC are applied. Dropping the short-time stationarity assumption of the voiced speech, this paper introduces the non-stationary signal analysis into the ASR framework. We present new acoustic features extracted by a pitch-adaptive Gammatone filter bank. The noise robustness was proved on AURORA 2 and 4 tasks, where the proposed features outperform the standard MFCC. Furthermore, successful combination experiments via ROVER indicate the differences between the new features and MFCC.

• Type: Journal
• Year: 2011
• Where: ICASSP
• Authors: Zoltán Tüske, Pavel Golik, Ralf Schlüter, Friedhelm R. Drepper
• Keywords: ICASSP 2011, Non-stationary Signal Analysis, Short-time Fourier Transform, Short-time Stationarity Assumption, Signal Processing