arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2002.08688eess.AScs.CVcs.LGcs.SD
收藏

Conv-TasNet的实证研究

An empirical study of Conv-TasNet

Dolby Laboratories(杜比实验室)

Berkan Kadioglu, Michael Horgan, Xiaoyu Liu, Jordi Pons, Dan Darcy, Vivek Kumar

摘要:

Conv-TasNet是最近提出的基于波形的深度神经网络,在语音源分离方面实现了最先进的性能。它的架构由一个可学习的编码器/解码器和一个在这个学习空间之上操作的分离器组成。已经对Conv-TasNet提出了各种改进。然而,他们主要关注分隔符,将其编码器/解码器作为(浅)线性运算符。在本文中,我们对Conv-TasNet进行了实证研究,并提出了一种基于其(深度)非线性变体的编码器/解码器增强。此外,我们对更大,更多样化的LibriTTS数据集进行了实验,并调查了在更大的数据集上训练时所研究模型的泛化能力。我们提出了跨数据集评估,包括评估WSJ 0 - 2 mix,LibriTTS和VCTK数据库的分离。我们的研究结果表明,增强的编码器/解码器可以提高平均SI-SNR性能超过1 dB。此外,我们还深入了解了Conv-TasNet的泛化能力以及编码器/解码器改进的潜在价值。

英文摘要:

Conv-TasNet is a recently proposed waveform-based deep neural network that achieves state-of-the-art performance in speech source separation. Its architecture consists of a learnable encoder/decoder and a separator that operates on top of this learned space. Various improvements have been proposed to Conv-TasNet. However, they mostly focus on the separator, leaving its encoder/decoder as a (shallow) linear operator. In this paper, we conduct an empirical study of Conv-TasNet and propose an enhancement to the encoder/decoder that is based on a (deep) non-linear variant of it. In addition, we experiment with the larger and more diverse LibriTTS dataset and investigate the generalization capabilities of the studied models when trained on a much larger dataset. We propose cross-dataset evaluation that includes assessing separations from the WSJ0-2mix, LibriTTS and VCTK databases. Our results show that enhancements to the encoder/decoder can improve average SI-SNR performance by more than 1 dB. Furthermore, we offer insights into the generalization capabilities of Conv-TasNet and the potential value of improvements to the encoder/decoder.

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。

↑