arXiv 2008.031492020-08-06eess.AScs.SD

基于辅助身份丢失的多级精细双径深度BiLSTM语音分离

Speech Separation Based on Multi-Stage Elaborated Dual-Path Deep BiLSTM with Auxiliary Identity Loss

Fujitsu Research and Development Center（富士通研发中心）；Harbin Institute of Technology（哈尔滨工业大学）

Ziqiang Shi, Rujie Liu, Jiqing Han

备注：: To appear in Interspeech 2020. arXiv admin note: substantial text overlap with arXiv:2001.08998, arXiv:1902.04891, arXiv:1902.00651

摘要：

具有双路径双向长短期记忆（BiLSTM）块的深度神经网络已被证明在序列建模中非常有效，特别是在语音分离中。这项工作研究了如何扩展双路径BiLSTM，以产生一种新的最先进的方法，称为TasTas，用于多说话者单声道语音分离（也称为鸡尾酒会问题）。TasTas介绍了两个简单但有效的改进，一个是迭代的多级细化方案，另一个是通过分离语音和原始语音之间的说话人身份一致性损失来校正具有不完美分离的语音，以提高基于双路径BiLSTM的网络的性能。TasTas将两个说话者的混合话语映射为两个单独的话语，其中每个话语仅包含一个说话者的声音。在WSJ 0 - 2 mix数据集上的实验结果表明，该网络在说话人分离方面有较大的性能提升，SDR提高了20.55dB，SI-SDR提高了20.35dB，PESQ提高了3.69，ESTOI提高了94.86%.我们在这里开源了我们对DPRNN-TasNet的重新实现（https：//github.com/Shiziqiang/dual-path-RNNs-DPRNNs-based-speech-separation），我们的TasTas是基于DPRNN-TasNet的这个实现的，相信本文的结果可以很容易地复制。

英文摘要：

Deep neural network with dual-path bi-directional long short-term memory (BiLSTM) block has been proved to be very effective in sequence modeling, especially in speech separation. This work investigates how to extend dual-path BiLSTM to result in a new state-of-the-art approach, called TasTas, for multi-talker monaural speech separation (a.k.a cocktail party problem). TasTas introduces two simple but effective improvements, one is an iterative multi-stage refinement scheme, and the other is to correct the speech with imperfect separation through a loss of speaker identity consistency between the separated speech and original speech, to boost the performance of dual-path BiLSTM based networks. TasTas takes the mixed utterance of two speakers and maps it to two separated utterances, where each utterance contains only one speaker's voice. Our experiments on the notable benchmark WSJ0-2mix data corpus result in 20.55dB SDR improvement, 20.35dB SI-SDR improvement, 3.69 of PESQ, and 94.86\% of ESTOI, which shows that our proposed networks can lead to big performance improvement on the speaker separation task. We have open sourced our re-implementation of the DPRNN-TasNet here (https://github.com/ShiZiqiang/dual-path-RNNs-DPRNNs-based-speech-separation), and our TasTas is realized based on this implementation of DPRNN-TasNet, it is believed that the results in this paper can be reproduced with ease.

机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。