BriefGPT.xyz
Jul, 2017
应用循环神经网络编码词汇混淆网络以进行对话状态跟踪
EncodingWord Confusion Networks with Recurrent Neural Networks for Dialog State Tracking
HTML
PDF
Glorianna Jagfeld, Ngoc Thang Vu
TL;DR
本文介绍了一种用循环神经网络对单词混淆网络进行编码的新方法，该方法可以代表自动语音识别系统的丰富假设空间，在口语系统的对话状态跟踪任务中，编码混淆网络优于编码自动语音识别的最佳假设。
Abstract
This paper presents our novel method to encode
word confusion networks
, which can represent a rich
hypothesis space
of
automatic speech recogniti
→