Tomas Gänsler

A double-talk detector based on coherence

We address the problem of detecting double-talk in a full duplex transmission line. A new double-talk detector (DTD) based on measuring the similarity between the far- and near-end speech signals is proposed. The detector is block oriented and operates in the frequency domain where the similarity between the signals is measured by means of the coherence function. The coherence is estimated with a short sequence of data by exploiting the multiple window spectrum estimation technique. Theoretical evaluation and examples of its performance are presented. The proposed DTD operates accurately in a wide range of situations, i.e., a difference in speech levels and hybrid attenuations ranging from 0 to 20 dB.

Robust extended multidelay filter and double-talk detector for acoustic echo cancellation

We propose an integrated acoustic echo cancellation solution based on a novel class of efficient and robust adaptive algorithms in the frequency domain, the extended multidelay filter (EMDF). The approach is tailored to very long adaptive filters and highly auto-correlated input signals as they arise in wideband full-duplex audio applications. The EMDF algorithm allows an attractive tradeoff between the well-known multidelay filter and the recursive least-squares algorithm. It exhibits fast convergence, superior tracking capabilities of the signal statistics, and very low delay. The low computational complexity of the conventional frequency-domain adaptive algorithms can be maintained thanks to efficient fast realizations. We also show how this approach can be combined efficiently with a suitable double-talk detector (DTD). We consider a corresponding extension of a recently proposed DTD based on a normalized cross-correlation vector whose performance was shown to be superior compared to other DTDs based on the cross-correlation coefficient. Since the resulting DTD also has an EMDF structure it is easy to implement, and the fast realization also carries over to the DTD scheme. Moreover, as the robustness issue during double talk is particularly crucial for fast-converging algorithms, we apply the concept of robust statistics into our extended frequency-domain approach. Due to the robust generalization of the cost function leading to a so-called M-estimator, the algorithms become inherently less sensitive to outliers, i.e., short bursts that may be caused by inevitable detection failures of a DTD. The proposed structure is also well suited for an efficient generalization to the multichannel case

Influence of audio coding on stereophonic acoustic echo cancellation

Stereophonic acoustic echo cancellation has been found more difficult than echo cancellation in mono due to a high correlation between the two audio channels. Different methods to decorrelate the channels have been proposed so that the stereophonic echo canceller identifies the true echo paths and its convergence rate increases. It is shown that the use of a perceptual audio coder effectively reduces the correlation between the channels and thus convergence to the true echo paths is insured. Furthermore, in those frequency regions where the encoder introduced quantization noise which is below the global perceptual masking threshold, an extra amount of inaudible noise can be added to the channels. Thereby the channel correlation is further decreased and the solution is stabilized. In subband audio coders with high frequency resolution only minor modifications are needed in the decoder.

New insights into the stereophonic acoustic echo cancellation problem and an adaptive nonlinearity solution

We expand the knowledge regarding the problems of two-channel (or stereophonic) echo cancellation. The major difference between two-channel and the single-channel echo cancellation is the problem of nonunique solutions in the two-channel case. In previous work, this nonuniqueness problem has been linked to the coherence between the two incoming audio channels. One proven solution to this problem is to distort the signals with a nonlinear device. In this work, we present new theory that gives insight to the existing links between: (i) coherence and level of distortion, and (ii) coherence and achievable misalignment of the stereophonic echo canceler. Furthermore, we present an adaptive nonlinear device that incorporates this new knowledge in such a way that a pre-specified maximum misalignment is maintained while improving the perceived quality by minimizing the introduced distortion. Moreover, all the ideas presented can be generalized to the multichannel (>2) case.

A frequency-domain double-talk detector based on a normalized cross-correlation vector

A key component for hands-free, full-duplex, communication technology is the echo canceler. An echo canceler consists primarily ofan adaptive 4lter and a control device called the double-talk detector. We derive a test statistic based on a normalized cross-correlation vector for the multichannel frequency-domain adaptive algorithm. The advantages ofthis approach are: low computational complexity, simplicity ofimplementation, and better perf ormance than the classical Geigel algorithm. ? 2001 Elsevier Science B.V. All rights reserved.

The fast normalized cross-correlation double-talk detector

In this paper, we present a method for handling double-talk. This approach uses a robust fast recursive least-squares algorithm (FRLS) and the normalized cross-correlation double-talk detector (NCC DTD). The NCC DTD is developed into a fast version, called FNCC, by reusing computational results from the FRLS algorithm. The major advantage of this detector is that it is much less dependent on the actual echo path attenuation than, e.g., the Geigel DTD. This makes it much more suitable for the acoustic application.

A double-talk resistant subband echo canceller

Abstract A recursive transfer function estimation algorithm is presented and analyzed. Applications can be found in either line or acoustic echo cancellation. The proposed algorithm is robust against burst disturbances that are caused by detection misses of double-talk present at the output of the echo path. A frequency-domain technique is used and a robust score function is derived from a criterion that is valid in the application. Analysis of the robust algorithm shows that good performance is to be expected. The performance of the algorithm when operated on real-life speech data in a full duplex communication system is shown by examples. Double-talk detection misses are shown to be well handled by the robust algorithm; yet, convergence rate and variance efficiency are as high as that of a non-robust least-squares algorithm.

Stereophonic acoustic echo cancellation and two-channel adaptive filtering: an overview

Stereophonic acoustic echo cancellation (SAEC), has been studied since the early 1990s. Because of spatial realism, two-channel audio is now becoming more interesting in a number of hands-free applications such as teleconferencing, multi-participant desktop conferencing, and televideo gaming. From a theoretical point of view, SAEC differs considerably from traditional mono echo cancellation since the normal equation to be solved by the adaptive algorithm is singular. Of course, this also affects the practical considerations that have to be made when designing a SAEC system, e.g. choice of adaptive algorithm. This paper reviews the basic theory behind SAEC, solutions to the singularity problem, and a number of classical two-channel adaptive algorithms suited to the problem. Copyright

Estimation of single event-related potentials utilizing the Prony method

This paper deals with estimation of the waveform of a single event-related potential, sERP. An additive noise model is used for the measured signal and the SNR of the disturbed sERP is approximately 0 dB. The sERP is described by a series expansion where the basis functions are damped sinusoids. The fundamental basis function is estimated by the least squares Prony method, derived for colored noise. The performance of the Prony method for different forms of the power density spectrum of the noise is investigated. A white noise approximation can be used at low signal-to-noise (SNR). The basis functions change slowly but the waveform of the sERP may vary from one stimulus to another, thus the authors average a small number of correlation functions in order to increase the SNR. The method is evaluated by using measurements from four subjects and the results confirm the variability of the sERP.

Fast communication: A widely linear model for stereophonic acoustic echo cancellation

The stereophonic acoustic echo, due to the coupling between two loudspeakers and two microphones, can be modelled by a two-input/two-output system with real random variables. In this paper, we recast the problem as a single-input/single-output system with complex random variables, by using the widely linear (WL) model, and propose a new distortion method that fits well in this context. In order to illustrate the behavior of this scheme, the recursive least-squares (RLS)-dichotomous coordinate descent (DCD) algorithm is used. Experimental results indicate that the RLS-DCD algorithm represents an attractive choice for this application since it has good numerical features in terms of stability and complexity.