arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2203.17152cs.SDcs.CLeess.AS
收藏

基于目标特征的感知对比度拉伸语音增强

Perceptual Contrast Stretching on Target Feature for Speech Enhancement

Academia Sinica(中央研究院);Microsoft Corporation(微软公司)

Rong Chao, Cheng Yu, Szu-Wei Fu, Xugang Lu, Yu Tsao

摘要:

由于使用深度学习模型作为基础函数,语音增强(SE)性能得到了显著改善。在这里,我们提出了一个感知对比度拉伸(PCS)的方法,以进一步提高SE性能。PCS是基于临界频带重要性函数推导出来的,并应用于修正SE模型的目标。具体而言,目标特征的对比度基于感知重要性被拉伸,从而提高整体SE性能。与基于后处理的实现相比,将PCS并入训练阶段保留了性能并减少了在线计算。值得注意的是,PCS可以与不同的SE模型架构和训练标准相结合。此外,PCS不影响SE模型训练的因果关系或收敛性。在VoiceBank-DEMAND数据集上的实验结果表明,该方法在因果(PESQ得分= 3.07)和非因果(PESQ得分= 3.35)SE任务上都能达到最先进的性能。

英文摘要:

Speech enhancement (SE) performance has improved considerably owing to the use of deep learning models as a base function. Herein, we propose a perceptual contrast stretching (PCS) approach to further improve SE performance. The PCS is derived based on the critical band importance function and is applied to modify the targets of the SE model. Specifically, the contrast of target features is stretched based on perceptual importance, thereby improving the overall SE performance. Compared with post-processing-based implementations, incorporating PCS into the training phase preserves performance and reduces online computation. Notably, PCS can be combined with different SE model architectures and training criteria. Furthermore, PCS does not affect the causality or convergence of SE model training. Experimental results on the VoiceBank-DEMAND dataset show that the proposed method can achieve state-of-the-art performance on both causal (PESQ score = 3.07) and noncausal (PESQ score = 3.35) SE tasks.

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。

↑