This paper presents a two-microphone speech enhancer designed to remove noise in hands-free car kits. The algorithm, based on the magnitude squared coherence, uses speech correlation and noise decorrelation to separate speech from noise. The remaining correlated noise is reduced using cross-spectral subtraction. Particular attention is focused on the estimation of the different spectral densiti...

Spectral parameter filtering such as RASTA [5], RASTA-like bandpass filtering and high-pass filtering operates on temporal dynamics of spectral parameters, and has been effective method to reduce channel distortions. Temporal derivatives (delta and delta-delta coefficients, that have proved as robust representation) and spectral mean normalization [4] are also equivalent to filtering that rejec...

The use of feature enhancement techniques to obtain estimates of the clean parameters is a common approach for robust automatic speech recognition (ASR). However, the decoding algorithm typically ignores how accurate these estimates are. Uncertainty decoding methods incorporate this type of information. In this paper, we develop a formulation of the uncertainty decoding paradigm for Frequency F...

Noise reduction is a very meaningful but difficult task and it has been a subject of intense research in recent years. This paper introduces two popular noise reduction techniques and presents our simulation result of a noise reduction system. It is shown that the system reduces the noise almost completely while keeps the enhanced speech signal very similar to the original speech signal.

Unvoiced speech separation is an important and challenging problem that has not received much attention. We propose a CASA based approach to segregate unvoiced speech from nonspeech interference. As unvoiced speech does not contain periodic signals, we first remove the periodic portions of a mixture including voiced speech. With periodic components removed, the remaining interference becomes mo...

An improved version of the original parametric formulation of the generalized spectral subtraction method is presented in this study. The original formulation uses parameters that minimize the mean-square error (MSE) between the estimated and true speech spectral amplitudes. However, the MSE does not take into account any perceptual measure. We propose two new short-time spectral amplitude esti...

Even a modest degree of room reverberation can greatly increase the difficulty of Automatic Speech Recognition. We have observed large increases in speech recognition word error rates when using a far-field (3-6 feet) mic in a conference room, in comparison with recordings from headmounted mics. In this paper, we describe experiments with a proposed remedy based on the subtraction of an estimat...

In this paper we describe our initial efforts to make SPHINX, the CMU continuous speech recognition system, environmentally robust. Our work has two major goals: to enable SPHINX to adapt to changes in microphone and acoustical environment, and to improve the performance of SPHINX when it is trained and tested using a desk-top microphone. This talk will describe some of our work in acoustical p...

The aim of this work is to enable a noise-free time-domain speech signal to be reconstructed from a stream of MFCC vectors and fundamental frequency and voicing estimates, such as may be received in a distributed speech recognition system. To facilitate reconstruction, both a sinusoidal model and a source-filter model of speech are compared by listening tests and spectrogram analysis, with the ...

A scaling theory of local-field fluctuations and optical nonlinearities is developed for random metaldielectric composites near a percolation threshold. The theory predicts that in the optical and infrared spectral ranges the local fields are very inhomogeneous and consist of sharp peaks representing localized surface plasmons. The localization maps the Anderson localization problem described b...