arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1608.07076cs.CL
收藏

一种面向对话系统的上下文感知自然语言生成器

A Context-aware Natural Language Generator for Dialogue Systems

Charles University in Prague(布拉格查理大学)

Ondřej Dušek, Filip Jurčíček

摘要:

我们提出了一种新的自然语言生成系统的口语对话系统能够夹带(适应)用户的说话方式,提供上下文适当的反应。该生成器是基于递归神经网络和序列到序列的方法。它完全可以从包括先前上下文以及要生成的响应的数据中进行训练。我们表明,上下文感知生成器产生了显着的改进,在自动度量和人类成对偏好测试的基线。

英文摘要:

We present a novel natural language generation system for spoken dialogue systems capable of entraining (adapting) to users' way of speaking, providing contextually appropriate responses. The generator is based on recurrent neural networks and the sequence-to-sequence approach. It is fully trainable from data which include preceding context along with responses to be generated. We show that the context-aware generator yields significant improvements over the baseline in both automatic metrics and a human pairwise preference test.

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。

↑