In this paper, we present an algorithm for multi-channel acoustic echo cancellation for a high-fidelity audio reproduction system equipped with a microphone array for voice control. Since a key requirement for this type of systems is leaving the reference signals that drive the multiple loudspeakers unaltered, we propose an adaptive decorrelation procedure in the time domain before feeding these signals to the echo cancellers. This helps mitigate the so-called non-uniqueness problem arising from their high correlation. This approach offers several advantages: it is extendable to a varying number of channels, requires low computational complexity, and preserves the original reference content. The echo cancellation approach applied to the decorrelated reference signals is then based on STFT-domain robust adaptive methods that do not require double-talk detection. The combination of these two techniques makes our approach particularly desirable for multi-channel echo cancellation problems with very low signal-to-echo ratio.

In this paper, we consider methods for improving far-field speech recognition using dereverberation based on sparse multi-channel linear prediction. In particular, we extend successful methods based on nonconvex iteratively reweighted least squares, that look for a sparse desired speech signal in the short-term Fourier transform domain, by proposing sparsity promoting convex functions. Furthermore, we show how to improve performance by applying regularization into both the reweighted least squares and convex methods. We evaluate the methods using large scale simulations by mimicking the application scenarios of interest. The experiments show that the proposed convex formulations and regularization offer improvements over existing methods with added robustness and flexibility in fairly different acoustic scenarios.

This paper presents a robust acoustic echo cancellation (AEC) system in the short-time Fourier transform (STFT) domain using adaptive crossband filters. The STFT-domain AEC allows for a simpler system structure compared to the traditional frequency-domain AEC, which normally requires several applications of the discrete Fourier transform (DFT) and the inverse DFT, while the robust AEC (RAEC) allows for continuous and stable filter updates during double talk without freezing the adaptive filter. The RAEC and the STFT-domain AEC have been investigated in the past in separate studies. In this work we propose a novel algorithm that combines the advantages of both approaches for robust update of the adaptive crossband filters even during double talk. Experimental results confirm the benefit of incorporating the robustness constraint for the adaptive crossband filters and show improved performance in terms of the echo reduction and the predicted sound quality.

In linear prediction of speech, the 1-norm error minimization criterion has been shown to provide a valid alternative to the 2-norm minimization criterion. However, unlike 2-norm minimization, 1-norm minimization does not guarantee the stability of the corresponding all-pole filter and can generate saturations when this is used to synthesize speech. In this paper, we introduce two new methods to obtain intrinsically stable predictors with the 1-norm minimization. The first method is based on constraining the roots of the predictor to lie within the unit circle by reducing the numerical range of the shift operator associated with the particular prediction problem considered. The second method uses the alternative Cauchy bound to impose a convex constraint on the predictor in the 1-norm error minimization. These methods are compared with two existing methods: the Burg method, based on the 1-norm minimization of the forward and backward prediction error, and the iteratively reweighted 2-norm minimization known to converge to the 1-norm minimization with an appropriate selection of weights. The evaluation gives proof of the effectiveness of the new methods, performing as well as unconstrained 1-norm based linear prediction for modeling and coding of speech.

Employing sparsity criteria in linear prediction of speech has been proven successful for several analysis and coding purposes. However, sparse linear prediction comes at the expenses of a much higher computational burden and numerical sensitivity compared to the traditional minimum variance approach. This makes sparse linear prediction difficult to deploy in real-time systems. In this paper, we present a step towards real-time implementation of the sparse linear prediction problem using hand-tailored interior-point methods. Using compiled implementations the sparse linear prediction problems corresponding to a frame size of 20ms can be solved on a standard PC in approximately 2ms and orders faster than with general purpose software.

The aim of this paper is to provide an overview of Sparse Linear Prediction, a set of speech processing tools created by introducing sparsity constraints into the linear prediction framework. These tools have shown to be effective in several issues related to modeling and coding of speech signals. For speech analysis, we provide predictors that are accurate in modeling the speech production process and overcome problems related to traditional linear prediction. In particular, the predictors obtained offer a more effective decoupling of the vocal tract transfer function and its underlying excitation, making it a very efficient method for the analysis of voiced speech. For speech coding, we provide predictors that shape the residual according to the characteristics of the sparse encoding techniques resulting in more straightforward coding strategies. Furthermore, encouraged by the promising application of compressed sensing in signal compression, we investigate its formulation and application to sparse linear predictive coding. The proposed estimators are all solutions to convex optimization problems, which can be solved efficiently and reliably using, e.g., interior-point methods. Extensive experimental results are provided to support the effectiveness of the proposed methods, showing the improvements over traditional linear prediction in both speech analysis and coding.

Linear prediction of speech based on 1-norm minimization has already proved to be an interesting alternative to 2-norm minimization. In particular, choosing the 1-norm as a convex relaxation of the 0-norm, the corresponding linear prediction model offers a sparser residual better suited for coding applications. In this paper, we propose a new speech modeling technique based on reweighted 1-norm minimization. The purpose of the reweighted scheme is to overcome the mismatch between 0-norm minimization and 1-norm minimization while keeping the problem solvable with convex estimation tools. Experimental results prove the effectiveness of the reweighted 1-norm minimization, offering better coding properties compared to 1-norm minimization.

In this paper, we describe a new approach to cope with packet loss in speech coders. The idea is to split the information present in each speech packet into two components, one to independently decode the given speech frame and one to enhance it by exploiting inter-frame dependencies. The scheme is based on sparse linear prediction and a redefinition of the analysis-by-synthesis process. We present Mean Opinion Scores for the presented coder with different degrees of packet loss and show that it performs similarly to frame dependent coders for low packet loss probability and similarly to frame independent coders for high packet loss probability. We also present ideas on how to make the coder work synergistically with the channel loss estimate.

Encouraged by the promising application of compressed sensing in signal compression, we investigate its formulation and application in the context of speech coding based on sparse linear prediction. In particular, a compressed sensing method can be devised to compute a sparse approximation of speech in the residual domain when sparse linear prediction is involved. We compare the method of computing a sparse prediction residual with the optimal technique based on an exhaustive search of the possible nonzero locations and the well known Multi-Pulse Excitation, the first encoding technique to introduce the sparsity concept in speech coding. Experimental results demonstrate the potential of compressed sensing in speech coding techniques, offering high perceptual quality with a very sparse approximated prediction residual.

In low bit-rate coders, the near-sample and far-sample redundancies of the speech signal are usually removed by a cascade of a short-term and a long-term linear predictor. These two predictors are usually found in a sequential and therefore suboptimal approach. In this paper we propose an analysis model that jointly finds the two predictors by adding a regularization term in the minimization process to impose sparsity constraints on a high order predictor. The result is a linear predictor that can be easily factorized into the short-term and long-term predictors. This estimation method is then incorporated into an algebraic code excited linear prediction scheme and shows to have a better performance than traditional cascade methods and other joint optimization methods, offering lower distortion and higher perceptual speech quality.

This paper presents two new classes of linear prediction schemes. The first one is based on the concept of creating a sparse residual rather than a minimum variance one, which will allow a more efficient quantization; we will show that this works well in presence of voiced speech, where the excitation can be represented by an impulse train, and creates a sparser residual in the case of unvoiced speech. The second class aims at finding sparse prediction coefficients; interesting results can be seen applying it to the joint estimation of long-term and short-term predictors. The proposed estimators are all solutions to convex optimization problems, which can be solved efficiently and reliably using, e.g., interior-point methods. Index Terms: linear prediction, all-pole modeling, convex optimization 1.