Research on VoIP Acoustic Echo Cancelation Algorithm Based on Speex

An acoustic echo canceller (AEC) is often used to eliminate acoustic echoes generated in a hands-free communication system. AEC approximates the echo path using an adaptive filter and eliminates the acoustic echo by subtracting the echo replica generated using the filter from the observed signal. However, conventional adaptive algorithms for updating filters fail to estimate during double talk periods when both acoustic echo and near-end speech are observed. In recent years, it was shown that blind source separation (BSS) can be used for echo cancellation during double talk period. However, the convergence speed is slower than the convergence speed of the conventional method. In our previous research we proposed combining conventional AEC and BSS, where BSS was used as preprocessor of adaptive filter to perform echo cancellation. Since there is room for improvement in the echo cancellation performance in previous research, in this paper we propose a new approach using BSS alternative output as AEC input. Simulation results show that the proposed system is effective for echo path estimation when echo path fluctuation occurs during double talk period.

In this paper, a single channel Kalman filtering in the spectral domain is proposed for acoustic echo and noise cancellation (AENC). An augmented extended Kalman filter is used in this work which is suitable for processing general complex-valued nonlinear and nonstationary signals. The near-end speech and background noise are modelled to make the filter robust against double-talk and against noise. In this work, the approach for echo cancellation does not depend on double-talk detector (DTD) to model the echo path, when signals other than the echoed far-end signal are present. The performance evaluations are conducted over the NOIZEUS corpus. The experimental results demonstrate that the ideal Kalman filter (with near-end speech parameters) outperforms other enhancement methods of echo cancellation in terms of subjective and objective measures. The non-ideal Kalman filter (when near-end speech parameters are obtained by an EM algorithm) also shows promising results after the ideal Kalman filter.

An acoustic echo canceler (AEC) is often employed to remove the acoustic echoes generated in hands-free communication systems. The AEC cancels the acoustic echoes by approximating the echo-path with the use of an adaptive filter and subtracting the pseudo echoes generated by the filter from the observed signal. The conventional adaptive algorithm for updating the filter, however, fails in estimation of the echo-path during double-talk when both the acoustic echo and the near-end speech are observed. Recently it has been shown that the blind source separation (BSS) can be employed to perform the echo cancelation during the double-talk; however, the convergence speed is slower than that of the conventional method. In this paper, a system is proposed that combines the conventional AEC and BSS; essentially, BSS is employed as a preprocessor for the adaptive filter performing the echo cancelation. Simulation results show that the proposed system is effective for the echo-path estimation during the double-talk as well as during variation in the echo path.

This paper considers an adaptive echo canceller which can be used in tasks of acoustic or electrical echo cancellation. The solution of these tasks usually requires the usage of the adaptive filters with a large number of weights. This means that a lot of computational resources of hardware or software are required for the echo canceller implementation. The usage of the Fast (computationally efficient) Affine Projection (FAP) adaptive filtering algorithm ensures the adaptive filter functional efficiency and complexity tradeoff in the echo cancellation tasks. The algorithm consists of two groups of calculations: a fast calculation of the adaptive filter error and a fast calculation of the normalizing matrix and its inversion. The second group is based on the usage of some calculations adopted from the sliding window Fast Recursive Least Squares adaptive filtering algorithms. However these calculations are potentially unstable. The given paper presents a simplified stable version of the multichannel FAP algorithm. The simplification is achieved due to the diagonalization of the normalizing matrix that ensures a simple procedure of direct calculation of this matrix and its inversion. If the echo signal is noisy (and that what always happens in our real world) then the original and the simplified FAP algorithms demonstrate about the same performance. The simulation of the two-channel acoustic echo cancellers confirms this statement.

An offline single channel acoustic echo cancellation (AEC) scheme is proposed based on gradient based adaptive least mean squares (LMS) algorithm considering a major practical application of echo cancellation system for enhancing recorded echo corrupted speech data. The unavailability of a reference signal makes the problem of single channel adaptive echo cancellation to be extremely difficult to handle. Moreover, continuous feedback of the echo corrupted signal to the input microphone can significantly degrade the quality of the original speech signal and may even result in howling. In order to overcome these problems, in the proposed scheme, the delayed version of the echo corrupted speech signal is considered as a reference. An objective function is thus formulated and thereby a modified LMS update equation is derived, which is shown to converge to the optimum Wiener-Hopf solution. The performance of the proposed method is evaluated in terms of both subjective and objective measures via extensive experimentation on several real-life echo corrupted signals and very satisfactory performance is obtained.

This paper presents new algorithms for acoustic echo cancellation and noise reduction which use two (or possibly more) microphone signals. In contrast to the single microphone method the multimicrophone approach can exploit the spatial coherence properties of sound fields which arise from noise and reverberated speech. Besides the standardfir echo canceller the proposed algorithms comprise an adaptive filter to eliminate non coherent signal components. The combined system achieves better Erle than thefir echo canceller alone, attenuates ambient noise, dereverberates near end speech, and possibly leads to implementations with reduced complexity. The paper analyzes the acoustical properties of typical environments, presents the algorithms and experimental results.

An acoustic echo cancellation system is one of the most important breakthrough in the field of adaptive systems. Today acoustic echo cancellers (AEC) are an integral part of full duplex hands-free voice communication. Conventional echo cancellers use a linear model to represent the echo path. However many consumer devices include loud-speakers and power amplifiers that generate non-linear distortions.Non-linearity occurs due to the use of low cost electronic loud speakers, microphones and poorly designed enclosures in an AEC system. Non-linearity causes vibration and harmonic distortion and also degrades the speech quality. Double talk detector (DTD) is a key component of an AEC. A DTD is used to sense when the far end signal is corrupted by the near end speech. The DTD freezes the adaptation of model filter to prevent the divergence of the adaptive filter. Various authors have proposed different algorithms for double talk detection. Some of the most popular algorithms are Geigel algorithm, cross-correlation based DTD, normalized cross correlation based DTD, variable impulse response DTD etc. In this paper several double talk detection algorithms in a non-linear platform of an AEC has been discussed.

This article explains the problem of acoustical echoes and their cancellation. It focuses on the hands‐free telephone as one of the major applications for echo cancelling devices. Beginning with a short discussion of a system formed by a loudspeaker and a microphone located within the same enclosure, the properties of speech signals and noise are described. The major part of the echo can be cancelled by an adaptive filter connected in parallel to the loudspeaker and the microphone. Residual echoes are suppressed by an additional filter within the outgoing signal path. Algorithms for the adaptation of the echo cancelling filter are described. Because of its robustness and its low computational complexity the NLMS algorithm is primarily applied. Measures to improve the speed of convergence and to avoid divergence in case of double talk or strong local noise are discussed. Echo cancellation for stereophonic systems, cancellation in subbands, and the applications of block processing techniques conclude the article.

Acoustic echo cancellation for teleconferencing or hands-free telephony constitutes a challenging task for today's digital signal processing techniques. Compared to the well-established line echo cancellation in telephone networks, the problem takes on a considerably larger size: The impulse response of the echo path to be compensated is longer and may vary more rapidly in time. Moreover, larger bandwidths are often desirable. Therefore, acoustic echo cancellation requires algorithms that adapt far more coeffecients and converge faster than those commonly used in line echo cancelers. On the other hand, direct implementation of sufficiently advanced adaptive filtering algorithms is usually prohibited because of their computational complexity and inherent numerical difficulties. As a way out, the subband approach realizes a “divide and conquer” strategy and promises a reduction of computational complexity and favorable circumstances for fast convergence at the cost of some extra delay. After comparing several possible subband structures, the frequency subband structure is discussed in some detail. For adaptation using LMS-type algorithms, reasons for the improved convergence behavior compared to a fullband implementation are explained. Design examples are presented to illustrate the effectiveness of the method.

One of the main difficulties in echo cancellation is the fact that the learning rate needs to vary according to conditions such as double-talk and echo path change. Several methods have been proposed to vary the learning. In this paper we propose a new closed-loop method where the learning rate is proportional to a misalignment parameter, which is in turn estimated based on a gradient adaptive approach. The method is presented in the context of a multidelay block frequency domain (MDF) echo canceller. We demonstrate that the proposed algorithm outperforms current popular double-talk detection techniques by up to 6 dB.

In hands-free telephone or video conference application, there exists an acoustic feedback coupling between the loudspeaker and microphone in an enclosed environment, which creates the acoustic echo. FIR filters are commonly used in acoustic echo cancellers because of their simple structure. However, in this paper, the Kautz and Laguerre filter structures are shown to be more efficient echo cancellers than the FIR filters, because they can describe accurately the acoustic echo system with fewer parameters. These filters are built from their respective orthonormal Kautz and Laguerre basis functions. The proposal is motivated by some theoretical and numerical results that the time-varying acoustic echo path is basically due to its time-varying zeros and not its time-invariant acoustical poles. Therefore, the poles of the Kautz and the Laguerre filters are estimated, and can be kept fixed or updated occasionally if required. The poles are estimated by a batch Gauss-Newton algorithm. Then, the coefficients of the Kautz and Laguerre filters can be estimated by most recursive algorithms that are suitable for linear regression models, e.g., the normalized LMS algorithm. Generally, it is shown that the proposed Kautz and Laguerre filters, as the filter structures in an acoustic echo canceller, have better convergence and tracking properties than the FIR and IIR filters.

When we talk with hands-free in a car or noisy lobby, the performance of the echo canceller degrade because background noise added to echo caused by the distance from mouth to microphone is relatively long. It gives a reason for necessity of noise-robust and high convergence speed adaptive algorithm. And if acoustic echo canceller operated not perfectly, residual signal going through the echo canceller to far-end speaker remains residual echo, which degrade quality of talk. To solve this problem, post-processing needed to remove residual echo ones more. In this paper, we propose a new acoustic echo canceller, which has noise robust and high convergence speed adaptive estimating echo path algorithm, linked with linear predictor and Wiener filter as a post-processor. By computer simulation, it is confirmed that the proposed algorithm shows better performance from acoustic interference cancellation (AIC) viewpoint.KeywordsNoise Robust AlgorithmPost-ProcessingLinear Prediction Error FilterWiener filter

This paper deals with the problem of echo cancellation of speech signals in an acoustic environment. In this regard, generally, different adaptive filter algorithms are employed, which may lack the flexibility of controlling the convergence rate, number of iterations, range of variation of filter coefficients, and tolerance consistency. In order to overcome these problems, unlike conventional approaches, we formulate the task of echo cancellation as a coefficient optimization problem whereby we introduce the particle swarm optimization (PSO) algorithm. In this case, the PSO is designed to perform the error minimization in frequency domain. From extensive experimentations, it is shown that the proposed PSO based acoustic echo cancellation method provides high echo cancellation performance in terms of echo return loss enhancement with a faster convergence rate in comparison to that obtained by some of the state-of-the-art methods.

Publication in the conference proceedings of EUSIPCO, Poznan, Poland, 2007

For convenient hands-free telecommunication systems echo cancelers with a low delay are desirable. Since such systems have to cope with long room impulse responses (RIRs) common echo cancelation techniques often suffer from slow convergence or introduce large delays. An algorithm for acoustic echo cancelation (AEC) has to compensate the influence of RIRs of several thousand sampling points. Thus for frequency domain algorithms long DFT-lengths are needed to cover the whole influence of the RIRs. Approaches in the partitioned frequency domain overcome this drawback. Our contribution describes a combined echo cancelation system in the partitioned frequency domain containing a conventional stereo AEC enhanced by a post-filter which requires a reliable residual echo estimate. This contribution focuses on minimizing the estimation bias introduced by partitioned residual echo estimation. For this purpose an optimal smoothing factor for the residual echo differential system will be derived.