This study proposes a permutation-alignment method for frequency-domain blind source separation. This method refines the results of Sawada’s permutation-alignment method using stricter objective function, relaxing the constraint of the permutation matrix by that of the doubly-stochastic matrix, and applying manifold optimization over it.

This paper proposes formulating declipping as a constrained multiple measurement vector (MMV) optimization problem that has a ${\ell _{2,0}}$ group norm as its cost function for further improving the state-of-the-art declipping method SParse Audio DEclipper (SPADE). This paper shows that the MMV optimization problem can be solved by extending the steps of alternating direction method of multipliers (ADMM) in the SPADE algorithm. The proposed method improved the signal-to-distortion ratio and reduced the cepstral distance for all clipping levels in a numerical simulation.

We propose a frequency-domain blind source separation method that uses (a) the ℓ1norm of orthonormal vectors of estimated source signals as a sparsity measure and (b) Cayley transform for optimizing the objective function under the unitary constraint in the Riemannian geometry approach. The orthonormal vectors of estimated source signals, obtained by the sphering of observed mixed signals and the unitary constraint on the separation filters, enables us to use the ℓ1norm properly as a sparsity measure. The Cayley transform enables us to handle the geometrical aspects of the unitary constraint efficiently. According to the simulation of a two-channel case, the proposed method achieved a 20-dB improvement in the source-to-interference ratio in a room with a reverberation time of T60= 300ms.

For reducing residual crosstalk in the output of blind source separation, we propose a frequency-domain post-filtering method that uses a multi-delay model of complex-valued residual crosstalk and sparsifies the estimates of the source signals. We formulate the reduction of residual crosstalk as an optimization problem using ℓ1norm and solve it using the alternating direction method of multiplier. The proposed method improved the source-to-interference ratio from 17.8 to 20.5 dB and the source-to-distortion ratio from 10.2 to 11.3 dB when it was combined with a brute-force solver of FastICA and reverberation time T60was 300 ms.

This article proposes a wave-domain residual echo reduction method for two-way immersive sound communication, which is based on wave field synthesis and uses uniform linear arrays of loudspeakers and microphones. This method is intended to improve the echo reduction performance in a mildly reverberant room with moderate increase in computational complexity. For improving such performance, this method takes residual echo due to reflected waves into account and uses a multi-tap echo-path model. To mitigate the increase in the method's computational complexity, it uses a size-reduced transfer function (TF) matrix instead of the full TF matrix. Specifically, this method extracts the dominant components of the loudspeaker signals adaptively by subspace tracking and estimates a size-reduced TF matrix that relates these dominant components to the residual echo. Simulation showed that the proposed method plus a wave-domain acoustic echo canceller (AEC) improved echo return loss enhancement by 10 dB compared to the conventional wave-domain echo reduction method plus the wave-domain AEC in a room with a reverberation time 180 ms. Its total computational complexity is twice that of the conventional echo reduction method plus a wave-domain AEC.

We propose a method to estimate a reverberant sound field by compressed sensing approach without using the hyper parameter for controlling sparsity. This method first applies sparse iterative covariance-based approach (SPICE) to the power spectral density matrix of the microphone signals and obtains the estimate of sensor noise power without using a hyper parameter. From this estimate, a convex optimization problem for a sparse solution is obtained that relates the microphone signals, plane-wave expansion coefficients, and the sensor noise power. The plane-wave expansion coefficients are obtained as the solution of this convex optimization problem. The sound field can be estimated from the plane-wave expansion coefficients.

We propose beamforming method that minimizes the ℓ1norm of a beamformer output vector under the same distortionless constraint as that of the conventional minimum power distortionless response (MPDR) beamformer. Using the ℓ1norm makes the beamformer output sparse. This leads to reducing the residual elements of the interference signal. In addition, the sensitivity of the proposed beamformer can be controlled by adding a norm constraint as in the MPDR beamformer. The proposed method improved the signal-to-interference-noise ratio by 7 dB from that of the MPDR beamformer for reverberation time T60= 300 ms in a simulation.

This paper proposes a new method to reduce residual echo for multichannel acoustic echo cancelers (AECs). This method estimates complex-valued residual echo and corrects its overestimated amplitude on the basis of a bias model of coherence estimation. Unlike conventional residual echo reduction methods for single-channel AECs, this method takes into account not only the magnitudes but also the phases of far-end signals and frequency responses. The inclusion of the phases gives the correct model of residual echo for multichannel AECs. The correction of the overestimated amplitude of residual echo decreases the distortion of near-end speech during double talk. The proposed method reduces the residual echo in the AEC output by 5-10 dB immediately after environmental changes such as a far-end talker change and an echo-path change. The proposed method provides less distorted near-end speech than the square-root Wiener filter during double talk.

We propose a method of estimating a sound field with two spherical microphone arrays (SMAs). This method estimates plane-wave expansion coefficients of the sound field by using sparse representation modeling in the frequency domain. The dictionary matrix we propose for this modeling achieves the integration of the measurements of two SMAs in a straightforward manner. The effectiveness of the proposed method was evaluated in simulations with computer-generated and measured impulse responses.

We propose a method for finding a good combination of nonidentical directivities for a microphone array, i.e., a combination that can achieve better noise suppression. To avoid checking all combinations of directivities, the method incrementally applies ℓ1-constrained minimum variance distortionless response beamforming to the coherently-focused power spectral density matrices of microphone array signals. The ℓ1constraint selects microphones with greater importance by inducing a sparse filter for beamforming.

Massive multichannel sound reproduction is necessary for highly immersive full-duplex communication. Efficient echo cancellation between massive loudspeakers and microphones is indispensable, and wave-domain adaptive filtering reduces the computational complexity drastically. We propose to further reduce this complexity by exploiting the band-limited property that the sound field has in the space-time-frequency representation. Experimental evaluations support the applicability of the proposed method.

We propose a posterior frequency-domain multi-channel residual echo canceling method. As the number of reproduction channels of an echo canceller increases, its adaptive filter shows a slower convergence of the filter coefficient error, that is, large echo path mismatch. In this situation, the change in the human talker in a remote site produces an abrupt residual echo more frequently. The proposed method immediately cancels this abrupt residual echo by estimating it from loudspeaker-reproduced signals, adaptive filter output, and echo replica with a much shorter frame length compared with adaptive filter length. By taking the echo replica into account, this frequency-domain method can improve residual-echo estimates. Another feature of the proposed method is controlling the residual echo estimates according to the confidence interval of these estimates.

We propose a novel step-size control method that treats the echo path imbalance problem in stereo echo cancellation. The imbalance between two echo paths in magnitude slows down the convergence rate of the adaptive filter for stereo echo cancellation. To speed up the convergence rate, the proposed method extracts the information about filter-coefficient errors of each unknown echo path, and controls the step-size of each channel accordingly.

A fast adaptive algorithm is required for stereo echo cancellation because the strong interchannel cross-correlation of stereo received signals leads to an ill-conditioned normal equation to be solved by the adaptive filter. An adaptive algorithm for this task should also be robust against disturbances such as near-end speech and near-end noise. We propose a coherence-based method of step-size control that provides robustness in stereo echo cancellation. Computer simulation demonstrates that the method is robust against near-end speech and noise, and is capable of quickly tracking changes in echo paths.

Highly cross-correlated input signals create the problem of slow convergence of misalignment in stereo echo cancellation even after undergoing non-linear preprocessing. We propose a new frequency-domain adaptive algorithm that improves the convergence rate by increasing the contribution of non-linearity in the adjustment vector. Computer simulation showed that it is effective when the non-linearity gain is small.

A computational approach is applied to the problem of controlling fixation of haptic perception during general motion. Estimation of hardness and curvatures of the environment from a single pressure image is shown to be ill-posed, and an active observation is demonstrated making the estimation well-posed. Also proposed is an algorithm for haptic servoing for the active finger-pad to fix its high-resolution region on the surface of contact.

Most algorithms currently proposed for sensor integration implicitly assume that the signals from all sensors are synchronized, but this does not hold in real situations. The paper proposes to handle time-variant sensor combination systems by a sequence, and formulates a method to verify that the designed sequence is stable, together with a method to predict theoretically the performance of the designed sequence based on the mutual information rate. This enables a sequence of sensor combinations to be designed rationally.