Integrating a Dialog Component into a Framework for Spoken Language Understanding
Authors: Weigelt, Sebastian, Hey, Tobias and Landhäußer, Mathias Conference: RAISE’18: IEEE/ACM 6th International Workshop on Realizing Artificial Intelligence Synergies in Software Engineering, 2018
Abstract: We develop PARSE, a framework for deep spoken language understanding. We use it for naturalistic end-user programming in spoken natural language. In this paper, we extend PARSE with a reactive dialog component to elicit additional user input and resolve natural language understanding issues. Dialog systems have been well studied but most applications are designed for dialog from the ground up - we see dialog primarily as a means to an end. Therefore, we integrated a dialog component with PARSE without affecting its other components: We modularize the dialog management and introduce Dialog Acts that bundle a trigger for the dialog and the reaction of the system. We implemented three Dialog Acts to address the following issues: speech recognition uncertainties, coreference ambiguities, and incomplete conditionals.We conducted a user study with ten subjects to evaluate our approach. The dialog component achieved resolution rates of 23% up to 50% (depending on the Dialog Act) and introduces a negligible number of errors. We expect the overall performance to increase even further with the implementation of additional Dialog Acts.
@inproceedings{Weigelt:2018:IDC:3194104.3194105,
author = {Weigelt, Sebastian and Hey, Tobias and Landh\"{a}u\sser, Mathias},
title = {Integrating a Dialog Component into a Framework for Spoken Language Understanding},
booktitle = {Proceedings of the 6th International Workshop on Realizing Artificial Intelligence Synergies in Software Engineering},
series = {RAISE '18},
year = {2018},
isbn = {978-1-4503-9358},
location = {Gothenburg, Sweden},
pages = {1--7},
numpages = {7},
url = {http://doi.acm.org/10.1145/3194104.3194105},
doi = {10.1145/3194104.3194105},
abstract = {Spoken language interfaces are the latest trend in human computer interaction. Users enjoy the newly found freedom but developers face an unfamiliar and daunting task. Creating reactive spoken language interfaces requires skills in natural language processing. We show how a developer can integrate a dialog component in a natural language processing system by means of software engineering methods. Our research project PARSE that aims at naturalistic end-user programming in spoken natural language serves as an example. We integrate a dialog component with PARSE without affecting its other components: We modularize the dialog management and introduce dialog acts that bundle a trigger for the dialog and the reaction of the system. We implemented three dialog acts to address the following issues: speech recognition uncertainties, coreference ambiguities, and incomplete conditionals. We conducted a user study with ten subjects to evaluate our approach. The dialog component achieved resolution rates from 23\% to 50\% (depending on the dialog act) and introduces a negligible number of errors. We expect the overall performance to increase even further with the implementation of additional dialog acts.},
acmid = {3194105},
publisher = {ACM},
address = {New York, NY, USA},
keywords = {dialog integration, dialog systems, enduser programming, human-computer interaction, knowledge-based software engineering, natural language processing for software engineering, naturalistic programming, programming in natural language},
}