arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2007.14598eess.AScs.SD
收藏

无参考DNN语音质量预测

DNN No-Reference PSTN Speech Quality Prediction

TU Berlin(柏林工业大学);Microsoft Corp.(微软公司)

Gabriel Mittag, Ross Cutler, Yasaman Hosseinkashi, Michael Revow, Sriram Srinivasan, Naglakshmi Chande, Robert Aichner

摘要:

传统的公共交换电话网(PSTN)对于VoIP网络提供商来说通常是一个黑匣子,因为它们无法访问性能指标,例如延迟或数据包丢失。只有退化的输出语音信号可以用来监测这些网络的语音质量。然而,当前最先进的语音质量模型不足以可靠地用于实时监测。其中一个原因是PSTN失真可能是唯一的,取决于提供商和国家,这使得很难训练一个模型,概括不同的PSTN网络。在本文中,我们提出了一个新的开源PSTN语音质量测试集,其中包含超过1000个众包真实电话呼叫。我们提出的无参考模型在验证和测试集上优于全参考POLQA和无参考P.563。此外,我们分析了文件裁剪对感知语音质量的影响,以及评级数量和训练大小对模型准确性的影响。

英文摘要:

Classic public switched telephone networks (PSTN) are often a black box for VoIP network providers, as they have no access to performance indicators, such as delay or packet loss. Only the degraded output speech signal can be used to monitor the speech quality of these networks. However, the current state-of-the-art speech quality models are not reliable enough to be used for live monitoring. One of the reasons for this is that PSTN distortions can be unique depending on the provider and country, which makes it difficult to train a model that generalizes well for different PSTN networks. In this paper, we present a new open-source PSTN speech quality test set with over 1000 crowdsourced real phone calls. Our proposed no-reference model outperforms the full-reference POLQA and no-reference P.563 on the validation and test set. Further, we analyzed the influence of file cropping on the perceived speech quality and the influence of the number of ratings and training size on the model accuracy.

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。

↑