Navin Chatlani

EMD-Based Filtering (EMDF) of Low-Frequency Noise for Speech Enhancement

An empirical mode decomposition-based filtering (EMDF) approach is presented as a postprocessing 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 log-spectral 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.

Adaptive Empirical Mode Decomposition for Signal Enhancement with application to speech

Speech enhancement is performed in a wide and varied range of instruments and systems. In this paper, a novel approach to signal enhancement using adaptive empirical mode decomposition (SEAEMD) is presented. Spectral analysis of non-stationary signals can be performed by employing techniques such as the STFT and the Wavelet transform, which use predefined basis functions. The empirical mode decomposition (EMD) performs very well in such environments and it decomposes a signal into a finite number of data-adaptive basis functions, called intrinsic mode functions (IMFs). The new SEAEMD system incorporates this multi-resolution approach with adaptive noise cancellation in order to perform signal enhancement on an IMF level. In comparison to the conventional adaptive noise cancellation, the application of SEAEMD to speech gives rise to improved quality and lower level of residual noise.

Single microphone wind noise PSD estimation using signal centroids

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.

Lower Arm Electromyography (EMG) Activity Detection Using Local Binary Patterns

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.

Improved pattern recognition classification accuracy for surface myoelectric signals using spectral enhancement

Abstract 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.

An alternative view of loudspeaker nonlinearities using the Hilbert-huang transform

This paper presents a new approach to nonlinear loudspeaker characterization using the Hilbert-Huang transform (HHT). Based upon the empirical mode decomposition (EMD) and the Hilbert transform, the HHT decomposes nonlinear signals into adaptive bases which reveal nonlinear effects in greater and more reliable detail than current approaches. Conventional signal decomposition techniques such as Fourier and Wavelet techniques analyse nonlinear distortion using linear transform theory. This restricts the nonlinear distortion to harmonic distortion. This work shows that real nonlinear loudspeaker distortion is more complex. HHT offers an alternate view through the cumulative effects of harmonics and intra-wave amplitude-and-frequency modulation. The work calls into question the interpretation of nonlinear distortion through harmonics and points towards a link between physical sources of nonlinearity and amplitude-and-frequency modulation. The work furthermore questions the suitability of traditional signal analysis approaches while giving weight to the use HHT analysis in future work.

Low complexity single microphone tonal noise reduction in vehicular traffic environments

A low complexity single microphone Tonal Noise Reduction (TNR) technique is presented for speech enhancement. This method is particularly effective in noisy environments which contain tonal noise sources, such as vehicular horns and alarms. TNR was designed to have low complexity and low memory requirements for use with battery operated communication devices. TNR detects the probability of the presence of these tonal noises which contaminate the desired speech signals. These noises are then attenuated using the proposed system for noise suppression. This is particularly effective for noise sources with a harmonic spectral structure. The proposed TNR system is able to maintain a balance between the level of noise reduction and speech distortion. Listening tests were performed to confirm the results. TNR can be used together with a general noise reduction system as a postprocessing stage by reducing the residual noise components.