The mismatch between the numerical and actual nonlinear models is a challenge to nonlinear acoustic echo cancellation (NAEC) when nonlinear adaptive filter is utilized. To alleviate this problem, we propose an effective method based on semi-blind source separation (SBSS), which uses a basis-generic expansion of the memoryless nonlinearity and then merges the unknown nonlinear expansion coefficients into the echo path. By regarding all the basis functions of the far-end input signal as the known equivalent reference signals, an SBSS updating algorithm is derived based on the constrained scaled natural gradient strategy. Unlike the commonly utilized adaptive algorithm, the proposed SBSS is based on the independence between the nearend signal and the reference signals, and is less sensitive to the mismatch of nonlinearity between the numerical and actual models. The experimental results with both simulated and real captured data validate the efficacy of the proposed method in NAEC.

The multichannel variational autoencoder (MVAE) integrates the rule-based update of a separation matrix and the deep generative model and proves to be a competitive speech separation method. However, the output (global) permutation ambiguity still exists and turns out to be a fundamental problem in applications. In this paper, we address this problem by employing two dedicated encoders. One encodes the speaker identity for the guidance of the output sorting, and the other encodes the linguistic information for the reconstruction of the source signals. The instance normalization (IN) and the adaptive instance normalization (adaIN) are applied to the networks to disentangle the speaker representations from the content representations. The separated sources are arranged in designated order by a symmetric permutation alignment scheme. In the experiments, we test the proposed method in different gender combinations and various reverberant conditions and generalize it to unseen speakers. The results validate its reliable sorting accuracy and good separation performance. The proposed method outperforms the other baseline methods and maintains stable performance, achieving over 20 dB SIR improvement even in high reverberant environments.

The complete decomposition performed by blind source separation is computationally demanding and superfluous when only the speech of one specific target speaker is desired. This letter proposes a computationally efficient blind source extraction method based on the fast fixed-point optimization algorithm under the mild assumption that the average power of the source of interest outweighs the interfering sources. Moreover, a one-unit scaling operation is designed to solve the scaling ambiguity for source extraction. Experiments validate the efficacy of the proposed method in extracting the dominant source.

The recently proposed semi-blind source separation (SBSS) method for nonlinear acoustic echo cancellation (NAEC) outperforms adaptive NAEC in attenuating the nonlinear acoustic echo. However, the multiplicative transfer function (MTF) approximation makes it unsuitable for real-time applications, especially in highly reverberant environments, and the natural gradient makes it hard to balance well between fast convergence speed and stability. In this paper, two more effective SBSS methods based on auxiliary-function-based independent vector analysis (AuxIVA) and independent low-rank matrix analysis (ILRMA) are proposed. The convolutive transfer function approximation is used instead of the MTF so that a long impulse response can be modeled with a short latency. The optimization schemes used in AuxIVA and ILRMA are carefully regularized according to the constrained demixing matrix of NAEC. The experimental results validate significantly better echo cancellation performances of the proposed methods.