Ali O. Abid Noor

Normalized Least Mean Square adaptive noise cancellation filtering for speaker verification in noisy environments

In this paper, we present a speaker verification system based on the hidden Markov models (HMMs) and normalized least mean square (NLMS) adaptive filtering. The aim of using NLMS adaptive filtering is to improve the HMMs performance in noisy environments. A Malay spoken digit database is used for the testing and validation modules. It is shown that, in a clean environment a total success rate (TSR) of 89.97% is achieved using HMMs. For speaker verification, the true speaker rejection rate is 25.3% while the impostor acceptance rate is 9.99% and the equal error rate (EER) is 16.66%. In noisy environments without NLMS adaptive filtering TSRs of between 43.07%-51.26% are achieved for SNRs of 0-30 dBs. Meanwhile, after NLMS filtering, TSRs of between 55.18%-55.30% are achieved for SNRs 0-30 dB.

Development of a Voice Activity Controlled Noise Canceller

In this paper, a variable threshold voice activity detector (VAD) is developed to control the operation of a two-sensor adaptive noise canceller (ANC). The VAD prohibits the reference input of the ANC from containing some strength of actual speech signal during adaptation periods. The novelty of this approach resides in using the residual output from the noise canceller to control the decisions made by the VAD. Thresholds of full-band energy and zero-crossing features are adjusted according to the residual output of the adaptive filter. Performance evaluation of the proposed approach is quoted in terms of signal to noise ratio improvements as well mean square error (MSE) convergence of the ANC. The new approach showed an improved noise cancellation performance when tested under several types of environmental noise. Furthermore, the computational power of the adaptive process is reduced since the output of the adaptive filter is efficiently calculated only during non-speech periods.

Adaptive Filtering Using Subband Processing: Application to Background Noise Cancellation

Adaptive filters are often involved in many applications, such as system identification, channel estimation, echo and noise cancellation in telecommunication systems. In this context, the Least Mean Square (LMS) algorithm is used to adapt a Finite Impulse Response (FIR) filter with a relatively low computation complexity and good performance. However, this solution suffers from significantly degraded performance with colored interfering signals, due to the large eigenvalue spread of the autocorrelation matrix of the input signal (Vaseghi, 2008). Furthermore, as the length of the filter is increased, the convergence rate of the algorithm decreases, and the computational requirements increase. This can be a problem in acoustic applications such as noise cancellation, which demand long adaptive filters to model the noise path. These issues are particularly important in hands free communications, where processing power must be kept as low as possible (Johnson et al., 2004). Several solutions have been proposed in literature to overcome or at least reduce these problems. A possible solution to reduce the complexity problem has been to use adaptive Infinite Impulse Response (IIR) filters, such that an effectively long impulse response can be achieved with relatively few filter coefficients (Martinez & Nakano 2008). The complexity advantages of adaptive IIR filters are well known. However, adaptive IIR filters have the well known problems of instability, local minima and phase distortion and they are not widely welcomed. An alternative approach to reduce the computational complexity of long adaptive FIR filters is to incorporate block updating strategies and frequency domain adaptive filtering (Narasimha 2007; Wasfy & Ranganathan, 2008). These techniques reduce the computational complexity, because the filter output and the adaptive weights are computed only after a large block of data has been accumulated. However, the application of such approaches introduces degradation in the performance, including a substantial signal path delay corresponding to one block length, as well as a reduction in the stable range of the algorithm step size. Therefore for nonstationary signals, the tracking performance of the block algorithms generally becomes worse (Lin et al., 2008). As far as speed of convergence is concerned, it has been suggested to use the Recursive Least Square (RLS) algorithm to speed up the adaptive process (Hoge et al., 2008).The convergence rate of the RLS algorithm is independent of the eigenvalue spread. Unfortunately, the drawbacks that are associated with RLS algorithm including its O(N2) computational requirements, which are still too high for many applications, where high

Interference control in speech using efficient subband LMS filtering

In this paper, an efficient interference canceller is developed using subband adaptive filtering. The aim is to reduce problems incorporated with the use of the conventional least mean square LMS adaptive algorithm in noise cancellation setup. The subband model is obtained by inserting a computationally efficient two fold oversampled filter banks in the conventional fullband model. The prototype filter of the filter bank is optimized for minimum amplitude distortion. Variable step-size LMS adaptive filters are used in subbands. Mean square error MSE convergence is used as a measure of performance under white and environmental noise. Compared to equivalent fullband and critically sampled systems, fast initial convergence is obtained. In addition to that, the amount of noise cancellation is improved by 10–15 dB on steady state.

Noise cancellation in speech using optimized subband adaptive filtering

In this paper, an optimized subband noise canceller for speech signals is developed. The canceller uses a two fold oversampled filter bank. The prototype filter of the filter bank is optimized for minimum system distortion. A variable step size least mean square LMS adaptive filter is used in each subband. Performance under white and colored environments is evaluated in terms of mean square error MSE performance. Comparison is made with a similar model that is based on a standard critically sampled filter banks. Fullband version is used as reference for the canceller outperformed both critically sampled and fullband models.

Development of efficient adaptive noise canceller using low complexity filter banks

In this paper an efficient adaptive noise canceller is presented. The canceller is based on low complexity optimized multirate filter banks. This work was motivated by the fact that adaptive noise cancellers suffer from slow convergence due to colored interfering signals as well as high computational burden due to long signal paths as in acoustic environment. The analysis stage is implemented with maximally decimated infinite impulse response IIR filter bank based on second order allpass sections. The synthesis filter bank is optimized in the input/output relationship leading to near perfect reconstruction and compensating for phase distortion. The proposed scheme offers a lower computational complexity than that offered by oversampled solution. The efficiency of the proposed canceller was compared to an equivalent system that uses over sampled finite impulse response FIR filter banks at the analysis and synthesis stages.

An approach for low power hearing aids design

A new low cost filterbank for hearing aids is presented. The analysis stage of the filterbank based on IIR allpass sections with modified response. The modification was performed by forcing phase nonlinearity at the analysis stage to be out of region of interest. By oversampling the filterbank by a factor of 2, the non-linear segments near the band edges was removed through subsequent synthesis filtering. Compared to existing literature designs, the new approach gave a substantial lowering in computational power, and a lower input/output delay.

Convergence improvement of the LMS adaptive noise canceller using low distortion filter banks

This paper presents a subband least mean square LMS noise canceller with improved convergence behaviour. This improvement is achieved by modifying and optimizing an existing multirate filter bank that is used to improve the performance of full-band LMS adaptive filters. The optimized oversampled subband noise canceller offers a simplified structure that without employing cross-filters or gap filter banks reduces the total input/output distortion in speech signals. Issues of increasing convergence speed and decreasing the residual noise at the system output are addressed. Performances under white and colored environments are evaluated experimentally in terms of mean square error MSE performance. Compared to an equivalent oversampled scheme, fast initial convergence and better noise reduction performance can be obtained with this approach.

An improved acoustic interference canceller using optimized IIR/FIR filter banks

In this paper an efficient adaptive noise canceller is presented. The canceller is based on low complexity infinite impulse response IIR filterbank at the analysis stage and an optimized finite impulse response FIR filterbank at the synthesis stage. The work is motivated by the fact that, the least mean square LMS adaptive noise cancellers suffer from slow convergence due to colored interfering signals, as well as high computational burden, due to long signal paths, as in acoustic environment. The efficiency of the proposed canceller was evaluated experimentally for white and colored interferences. The new scheme offers a lower computational complexity than that offered by oversampled solution with FIR filterbanks at analysis stage.