Measuring the productivity of an agent in a call center domain is a challenging task. Subjective measures are commonly used for evaluation in the current systems. In this paper, we propose an objective framework for modeling agent productivity for real estate call centers based on speech signal processing. The problem is formulated as a binary classification task using deep learning methods. We explore several designs for the classifier based on convolutional neural networks (CNNs), long-short-term memory networks (LSTMs), and an attention layer. The corpus consists of seven hours collected and annotated from three different call centers. The result shows that the speech-based approach can lead to significant improvements (1.57% absolute improvements) over a robust text baseline system.

中文翻译：

---

细心卷积神经网络的呼叫中心领域的高级生产力建模

衡量呼叫中心域中业务代表的生产率是一项艰巨的任务。在当前系统中，主观测量通常用于评估。在本文中，我们提出了一个基于语音信号处理的房地产呼叫中心座席生产率建模的客观框架。使用深度学习方法将问题表述为二进制分类任务。我们探索基于卷积神经网络（CNN），长期短期记忆网络（LSTM）和关注层的分类器设计。语料库由从三个不同的呼叫中心收集并注释的七个小时组成。结果表明，与健壮的文本基线系统相比，基于语音的方法可以带来显着改善（绝对改善1.57％）。