A Review of Signal Subspace Speech Enhancement and Its Application to Noise Robust Speech Recognition

Abstract

The objective of this paper is threefold: (1) to provide an extensive review of signal subspace speech enhancement, (2) to derive an upper bound for the performance of these techniques, and (3) to present a comprehensive study of the potential of subspace filtering to increase the robustness of automatic speech recognisers against stationary additive noise distortions. Subspace filtering methods are based on the orthogonal decomposition of the noisy speech observation space into a signal subspace and a noise subspace. This decomposition is possible under the assumption of a low-rank model for speech, and on the availability of an estimate of the noise correlation matrix. We present an extensive overview of the available estimators, and derive a theoretical estimator to experimentally assess an upper bound to the performance that can be achieved by any subspace-based method. Automatic speech recognition (ASR) experiments with noisy data demonstrate that subspace-based speech enhancement can significantly increase the robustness of these systems in additive coloured noise environments. Optimal performance is obtained only if no explicit rank reduction of the noisy Hankel matrix is performed. Although this strategy might increase the level of the residual noise, it reduces the risk of removing essential signal information for the recogniser's back end. Finally, it is also shown that subspace filtering compares favourably to the well-known spectral subtraction technique.

Speech Enhancement Based on Noise Eigenspace Projection

“…(9). On one hand, according to the projection theorem [16], the speech signal in a frame, x, can be represented using the speech eigenvector, ψ j .…”

“…They have been used widely because of their simplicity and high computational efficiency. Recently, signal subspace approaches [5]- [9] have been proposed for enhancing speech signals. The core idea of these signal subspace approaches is to decompose a noisy speech into uncorrelated components in a signal space.…”

Speech Denoising in White Noise Based on Signal Subspace Low-rank Plus Sparse Decomposition

“…The typical algorithms including spectral subtraction [1], minimum mean square error (MMSE) estimation [2][3][4], Wiener filtering [5][6][7][8], and subspace methods [9][10][11][12][13]. Spectral Subtraction and Wiener filtering have been widely used for enhancing speech because of their simplicity and ease of implementation in single channel systems but they suffer from the production of musical noise after enhancement and is one of their major drawbacks.…”

“…Spectral Subtraction and Wiener filtering have been widely used for enhancing speech because of their simplicity and ease of implementation in single channel systems but they suffer from the production of musical noise after enhancement and is one of their major drawbacks. Signal subspace approach [9][10][11][12][13], have shown to give a better compromise between less residual noise and signal distortion of the output signal, compared to the other existing techniques.…”

Binary Mask Estimation using Training-based SNR Estimation for Improving Speech Intelligibility

“…지금까지 시도되었던 잡 음 제거 기법들은 주로 잡음 환경에서의 음성 통신을 위해 사용자가 좀 더 편안하게 들을 수 있도록 음질을 개선하는데 초점이 맞춰져 있다 [1][2][3][4] . 또한, 음성 인식에 대한 연구가 활발 해지고 음성 인식의 실용화를 위해 생활 잡음 환경에서 취득 된 음성 신호의 잡음을 억제하고자하는 연구들도 많이 진행 되어 왔다 [5,6] . 이때까지의 연구들을 통해 잡음을 제거하여 음질 (quality)을 개선하거나 음성 인식의 성능을 높여왔지 만 주로 시간에 따라 특성이 빨리 변하지 않는 잡음, 즉 정상 잡음 (stationary noise)에 국한되어 왔다.…”