An Empirical Mode Decomposition based filtering (EMDF) approach is presented as a post-processing stage for speech enhancement. This method is particularly effective in low frequency noise environments. Unlike previous EMD based denoising methods, this approach does not make the assumption that the contaminating noise signal is fractional Gaussian Noise. An adaptive method is developed to select the IMF index for separating the noise components from the speech based on the second-order IMF statistics. The low frequency noise components are then separated by a partial reconstruction from the IMFs. It is shown that the proposed EMDF technique is able to suppress residual noise from speech signals that were enhanced by the conventional optimally-modified logspectral amplitude approach which uses a minimum statistics based noise estimate. A comparative performance study is included that demonstrates the effectiveness of the EMDF system in various noise environments, such as car interior noise, military vehicle noise and babble noise. In particular, improvements up to 10 dB are obtained in car noise environments. Listening tests were performed that confirm the results.

This paper presents a new electromyography activity detection technique in which 1-D local binary pattern histograms are used to distinguish between periods of activity and inactivity in myoelectric signals. The algorithm is tested on forearm surface myoelectric signals occurring due to hand gestures. The novel features of the presented method are that: 1) activity detection is performed across multiple channels using few parameters and without the need for majority vote mechanisms, 2) there are no per-channel thresholds to be tuned, which makes the process of activity detection easier and simpler to implement and less prone to errors, 3) it is not necessary to measure the properties of the signal during a quiescent period before using the algorithm. The algorithm is compared to other offline single- and double-threshold activity detection methods and, for the data sets tested, it is shown to have a better overall performance with greater tolerance to the noise in the real data set used.

This contribution presents an efficient technique for the enhancement of speech signals disturbed by wind noise. In almost all noise reduction systems an estimate of the current noise power spectral density (PSD) is required. As common methods for background noise estimation fail due to the non-stationary characteristics of wind noise signals, special algorithms are required. The proposed estimation technique consists of three steps: a feature extraction followed by a wind noise detection and the calculation of the current wind noise PSD. For all steps we exploit the different spectral energy distributions of speech and wind noise. In this context, the so-called signal centroids are introduced. Investigations with measured audio data show that our method can cope with the non-stationary characteristics and enables a sufficient reduction of wind noise. In contrast to other wind noise reduction schemes the proposed algorithm has low complexity and low memory consumption.

In this paper, we demonstrate that Spectral Enhancement techniques can be configured to improve the classification accuracy of a pattern recognition-based myoelectric control system. This is based on the observation that, when the subject is at rest, the power in EMG recordings drops to levels characteristic of the noise. Two Minimum Statistics techniques, which were developed for speech processing, are compared against electromyographic (EMG) de-noising methods such as wavelets and Empirical Mode Decomposition. In the cases of simulated EMG signals contaminated with white noise and for real EMG signals with added and intrinsic noise the gesture classification accuracy was shown to increase. The mean improvement in the classification accuracy is greatest when Improved Minima-Controlled Recursive Averaging (IMCRA)-based Spectral Enhancement is applied, thus demonstrating the potential of Spectral Enhancement techniques for improving the performance of pattern recognition-based myoelectric control

No abstract