arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2209.01702eess.AS
收藏

基于GAN的时域语音超分辨电话说话人确认建模

Time-domain speech super-resolution with GAN based modeling for telephony speaker verification

Saurabh Kataria, Jesús Villalba, Laureano Moro-Velázquez, Piotr Żelasko, Najim Dehak

摘要:

自动说话人验证(ASV)技术在虚拟助理中已经变得司空见惯。然而,当训练域和测试域之间不匹配时,其性能受到影响。混合带宽训练,即,汇集来自这两个域的训练数据是开发适用于窄带域和宽带域的通用模型的优选选择。我们建议通过对窄带信号进行神经上采样来补充这种技术,也称为带宽扩展。我们的主要目标是发现和分析基于高性能时域生成对抗网络(GAN)的模型,以改进我们下游最先进的ASV系统。我们选择GANs是因为它们(1)对于学习条件分布是强大的,(2)允许在训练下游任务(ASV)时灵活地使用插件作为预处理器。以前的工作主要集中在特征域的带宽扩展和有限的实验设置。我们通过1)使用时域扩展模型,2)报告三个真实测试集的结果,2)扩展训练数据,以及3)设计新的测试时间方案来解决这些限制。我们比较了有监督的(条件GAN)和无监督的GANs(循环GAN),并证明了平均相对改善相等错误率分别为8.6%和7.7%。为了进一步分析,我们研究了声谱图视觉质量,音频感知质量,t-SNE嵌入和ASV分数分布的变化。我们表明,我们的带宽扩展导致的现象,如移动电话(测试)嵌入宽带(训练)信号,感知质量与下游性能的负相关性,和条件无关的分数校准。

英文摘要:

Automatic Speaker Verification (ASV) technology has become commonplace in virtual assistants. However, its performance suffers when there is a mismatch between the train and test domains. Mixed bandwidth training, i.e., pooling training data from both domains, is a preferred choice for developing a universal model that works for both narrowband and wideband domains. We propose complementing this technique by performing neural upsampling of narrowband signals, also known as bandwidth extension. Our main goal is to discover and analyze high-performing time-domain Generative Adversarial Network (GAN) based models to improve our downstream state-of-the-art ASV system. We choose GANs since they (1) are powerful for learning conditional distribution and (2) allow flexible plug-in usage as a pre-processor during the training of downstream task (ASV) with data augmentation. Prior works mainly focus on feature-domain bandwidth extension and limited experimental setups. We address these limitations by 1) using time-domain extension models, 2) reporting results on three real test sets, 2) extending training data, and 3) devising new test-time schemes. We compare supervised (conditional GAN) and unsupervised GANs (CycleGAN) and demonstrate average relative improvement in Equal Error Rate of 8.6% and 7.7%, respectively. For further analysis, we study changes in spectrogram visual quality, audio perceptual quality, t-SNE embeddings, and ASV score distributions. We show that our bandwidth extension leads to phenomena such as a shift of telephone (test) embeddings towards wideband (train) signals, a negative correlation of perceptual quality with downstream performance, and condition-independent score calibration.

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。

↑