The model takes as input a sequence of dialogue contexts and outputs a sequence of dialogue acts corresponding to user intentions. The dialogue contexts include information about the machine acts and the status of the user goal.
Jun 30, 2016
In this paper, we examine simple sequence-to-sequence neural network architectures for training end-to-end, natural language to natural language, user ...
With the advent of deep-learning it is possible to train sequence-to-sequence (seq-2-seq) models as simulated users that track an external agenda [11]. It would ...
Aug 20, 2017 · In this paper, we examine simple sequence-to-sequence neural network architectures for training end-to-end, natural language to natural language ...
A data-driven user simulator based on an encoder-decoder recurrent neural network that outperforms an agenda-based simulator and an n-gram simulator.
The user and system dialogue acts can easily be mapped to those in DSTC2. We use this dataset in order to evaluate the user simulators on a new, larger domain.
Aug 20, 2017 · Unlike the above model that works on the semantic level of dialogue acts, sequence-to-sequence models have also been used to generate the text ...
User simulation is essential for generating enough data to train a statistical spoken dialogue system. Previous models for user simulation suffer from several ...
Sequence to Sequence Modeling for User Simulation in Dialog Systems. Crook P., Marin A. Expand. Publication type: Proceedings Article. Publication date: 2017 ...
In order to build user simulators, we need to model user behavior and therefore, we annotate the user-side dialog act in the restaurant domain of Multiwoz.