Acoustical and Environmental Robustness in Automatic Speech Recognition

Abstract

This dissertation describes a number of algorithms developed to increase the robustness of automatic speech recognition systems with respect to changes in the environment. These algorithms attempt to improve the recognition accuracy of speech recognition systems when they are trained and tested in different acoustical environments, and when a desk-top microphone (rather than a close-talking microphone) is used for speech input. Without such processing, mismatches between training and testing conditions produce an unacceptable degradation in recognition accuracy.Two kinds of environmental variability are introduced by the use of desk-top microphones and different training and testing conditions: additive noise and spectral tilt introduced by linear filtering. An important attribute of the novel compensation algorithms described in this thesis is that they provide joint rather than independent compensation for these two types of degradation.Acoustical compensation is applied in our algorithms as an additive correction in the cepstral domain. This allows a higher degree of integration within SPHINX, the Carnegie Mellon speech recognition system, that uses the cepstrum as its feature vector. Therefore, these algorithms can be implemented very efficiently. Processing in many of these algorithms is based on instantaneous signal-to-noise ratio (SNR), as the appropriate compensation represents a form of noise suppression at low SNRs and spectral equalization at high SNRs.The compensation vectors for additive noise and spectral transformations are estimated by minimizing the differences between speech feature vectors obtained from a "standard" training corpus of speech and feature vectors that represent the current acoustical environment. In our work this is accomplished by a minimizing the distortion of vector-quantized cepstra that are produced by the feature extraction module in SPHINX.In this dissertation we describe several algorithms including the SNR-Dependent Cepstral Normalization, (SDCN) and the Codeword-Dependent Cepstral Normalization (CDCN). With CDCN, the accuracy of SPHINX when trained on speech recorded with a close-talking microphone and tested on speech recorded with a desk-top microphone is essentially the same obtained when the system is trained and tested on speech from the desk-top microphone.An algorithm for frequency normalization has also been proposed in which the parameter of the bilinear transformation that is used by the signal-processing stage to produce frequency warping is adjusted for each new speaker and acoustical environment. The optimum value of this parameter is again chosen to minimize the vector-quantization distortion between the standard environment and the current one. In preliminary studies, use of this frequency normalization produced a moderate additional decrease in the observed error rate.

VTLN adaptation for statistical speech synthesis

“…Another stage of bilinear transform can be cascaded with the existing one to accommodate the VTLN warping. It has been shown [10] that the combination of two bilinear transforms with warping factors α1 and α2 is equivalent to a bilinear transform with single warping factor given by:…”

Signal adaptive spectral envelope estimation for robust speech recognition

“…A derivation of (24) is provided in [29]. To get a flat transfer function, we now apply the inverted weighting function…”

Towards non-stationary model-based noise adaptation for large vocabulary speech recognition

“…fŜ R g = argmax k l P (Y S k R l ) (6) where S k = fs 1 k s 2 k : : : s M k g is a particular speech state sequence and R l is a particular noise state sequence.…”

“…These techniques are called feature-based methods and model-based methods, respectively. Cepstral Mean Normalization [8] and Spectral Subtraction [9], and their extensions [6] are examples of feature-based methods.…”