Traditionally, the quality of acoustic echo cancellers is evaluated using intrusive speech quality assessment measures such as ERLE \cite{g168} and PESQ \cite{p862}, or by carrying out subjective laboratory tests. Unfortunately, the former are not well correlated with human subjective measures, while the latter are time and resource consuming to carry out. We provide a new tool for speech quality assessment for echo impairment which can be used to evaluate the performance of acoustic echo cancellers. More precisely, we develop a neural network model to evaluate call quality degradations in two separate categories: echo and degradations from other sources. We show that our model is accurate as measured by correlation with human subjective quality ratings. Our tool can be used effectively to stack rank echo cancellation models. AECMOS is being made publicly available as an Azure service.

翻译：传统上,对声响取消器的质量进行评估时,采用侵入性语言质量评估措施,如ERLE\cite{g168}和PESQ\cite{p862}等,或进行主观实验室测试。不幸的是,前者与人的主观措施不完全相关,而后者则耗费时间和资源。我们为回声损害的语音质量评估提供了一个新的工具,可用于评估声响取消器的性能。更确切地说,我们开发了一个神经网络模型,用于评估两个不同类别中调用的质量退化:来自其他来源的回声和退化。我们显示,我们的模型准确性与人的主观质量评级相关。我们的工具可以有效地用于堆叠回声取消模型。AECMOS正在作为一种Azure服务公开提供。