Compressed sensing (CS) is a sampled approach on signal sparsity-base, and it can effectively extract the information which is contained in the signal. This paper presents a noisy speech enhancement new method based on CS process. Algorithm uses a voice sparsity in the discrete fast Fourier transform (Fast Fourier transform, FFT), and complex domain observation matrix is designed, and the noisy...

In this paper, we evaluate a performance of in-car speech enhancement techniques with single channel signal processing and microphone array signal processing. We employ SS (Fourier Spectral Subtraction) and WSS (Wavelet Spectral Subtraction) as single channel signal processing and delay-and-sum beamformer, eigen beamformer, AMNOR (Adaptive Microphone array for NOise Reduction), S-AMNOR, Svc-AMN...

In this paper, we propose a feature enhancement algorithm for wireless speech recognition in adverse acoustic environments. A speech recognition system is realized at the network side of a wireless communications system and feature parameters are extracted directly from the bitstream of the speech coder employed in the system, where the feature parameters are composed of spectral envelope infor...

This paper presents a new speech enhancement approach originated from factor analysis (FA) framework. FA is a data analysis model where the relevant common factors can be extracted from observations. A factor loading matrix is found and a resulting model error is introduced for each observation. Interestingly, FA is a subspace approach properly representing the noisy speech. This approach parti...

In recent years, Automatic Speech Recognition (ASR) systems designed to work in controlled environments using clean speech have reached very high levels of performance. However, the accuracy of speech recognition degrades severely when the systems are operated in noisy environments. In this thesis we address the problem of single-channel speech enhancement. Starting with a study of the state-of...

The enhancement of speech degraded by nonstationary noises and low signal-to-noise ratio (SNR) conditions is a high demanding but challenging task. We present a robust and effective single channel speech enhancement algorithm under the framework of Nonnegative Matrix Factorization (NMF). Considering the sparse property of speech and low-rank property of nonstationary noise, a new formulation fo...

While the state-of-the-art speech enhancement methods are focused on the modification of the noisy spectral amplitude, our recent findings demonstrate positive impact of incorporating the speech phase spectrum in speech enhancement. In this show and tell proposal, we demonstrate the recent progress towards utilizing the phase information in closed-loop iterative manner leading to the joint enha...

A clean speech VQ codebook has been shown to be effective in imposing intraframe constraints in Iterative Wiener Filtering (CCIWF) for speech enhancement. However, for time-varying noises, the performance is sub-optimum. We propose a smoothed noise update technique that uses the estimated signal spectrum for subsequent signal estimation. This leads to a more effective solution than the soft-dec...