The Kalman filter is widely used for addressing acoustic echo cancellation (AEC) problems due to their robustness to double-talk and fast convergence. However, the inability to model nonlinearity and the need to tune control parameters cast limitations on such adaptive filtering algorithms. In this paper, we integrate the frequency domain Kalman filter (FDKF) and deep neural networks (DNNs) into a hybrid method, called KalmanNet, to leverage the advantages of deep learning and adaptive filtering algorithms. Specifically, we employ a DNN to estimate nonlinearly distorted far-end signals, a transition factor, and the nonlinear transition function in the state equation of the FDKF algorithm. Experimental results show that the proposed KalmanNet improves the performance of FDKF significantly and outperforms strong baseline methods.
翻译:Kalman过滤器被广泛用于解决声学回声取消(AEC)问题,原因是它们具有很强的双轨和快速趋同能力。然而,由于无法建模非线性和调控参数的必要性,对此类适应性过滤算法施加了限制。在本文中,我们将频率域域Kalman过滤器(FDKF)和深神经网络(DNNs)纳入称为KalmanNet的混合方法,以利用深层次学习和适应性过滤算法的优势。具体地说,我们使用DNN来估算非线性扭曲的远端信号、过渡系数以及FDKF算法状态方程式中的非线性过渡功能。实验结果表明,拟议的KalmanNet大大改进了FDKF的性能并超越了强大的基线方法。