双路注意力循环网络的轻量化语音分离

Light-weight speech separation based on dual-path attention and recurrent neural network

提出了双路注意力循环网络的轻量化语音分离方法。首先,该方法使用基于“双路注意力机制”和“双路循环网络”的可选择分支结构对语音信号进行建模,从而提取深层特征信息并降低模型的参数量。其次,引入子带处理技术,从而降低模型的计算量。在LibriCSS数据集上的实验结果表明,该方法取得的平均词错误率为8.6%,且参数量和计算量分别仅为0.15 MiB和15.2 G/6s,与当前主流方法相比,分别减小了3.3~391.3倍和1.1~3.2倍。这表明,所提方法在取得高语音分离性能的同时,能有效地降低模型的参数量和计算量。

A light-weight speech separation algorithm based on dual-path attention and recurrent neural network is proposed.First,optional branch structures based on dual-path attention mechanism and dual-path recurrent network are utilized to model the speech signals,which facilitate the extraction of deep feature information and the reduction of training parameters.Second,sub-band processing approach is introduced to alleviate the computation burden.As shown by the experimental results on the LibriCSS dataset,the average word error rate obtained by the proposed algorithm is 8.6%with only 0.15 MiB training parameters and 15.2 G/6s computation cost,which is 3.3−391.3 and 1.1−3.2 times smaller than other mainstream approaches.This proves the proposed algorithm can effectively reduce the training parameters and computation cost while achieving high speech separation performance.