Computing MMSE Estimates and Residual Uncertainty Directly in the Feature Domain of ASR using STFT Domain Speech Distortion Models

• Published in: IEEE transactions on audio, speech, and language processing speech, and language processing, 2013-05, Vol.21 (5), p.1023-1034
• Main Authors: Astudillo, R. F., Orglmeister, R.
• Format: Article
• Language: English
• Subjects: Applied sciences; Coding, codes; Computation; Computational modeling; Detection, estimation, filtering, equalization, prediction; Distortion; Estimates; Estimators; Exact sciences and technology; Feature extraction; Information, signal and communications theory; Mathematical models; Mean square errors; Mel frequency cepstral coefficient; MMSE; Propagation; Robustness; Signal and communications theory; Signal processing; Signal, noise; Speech; Speech enhancement; Speech processing; Studies; Telecommunications and information theory; Uncertainty; uncertainty decoding; uncertainty propagation; wiener filter
• ISSN: 1558-7916; 2329-9290; 1558-7924; 2329-9304
• DOI: 10.1109/TASL.2013.2244085

In this paper we demonstrate how uncertainty propagation allows the computation of minimum mean square error (MMSE) estimates in the feature domain for various feature extraction methods using short-time Fourier transform (STFT) domain distortion models. In addition to this, a measure of estimate reliability is also attained which allows either feature re-estimation or the dynamic compensation of automatic speech recognition (ASR) models. The proposed method transforms the posterior distribution associated to a Wiener filter through the feature extraction using the STFT Uncertainty Propagation formulas. It is also shown that non-linear estimators in the STFT domain like the Ephraim-Malah filters can be seen as special cases of a propagation of the Wiener posterior. The method is illustrated by developing two MMSE-Mel-frequency Cepstral Coefficient (MFCC) estimators and combining them with observation uncertainty techniques. We discuss similarities with other MMSE-MFCC estimators and show how the proposed approach outperforms conventional MMSE estimators in the STFT domain on the AURORA4 robust ASR task.