Publication Type

Conference Proceeding Article

Version

publishedVersion

Publication Date

4-2015

Abstract

Language barrier is the primary challenge for effectivecross-lingual conversations. Spoken language translation(SLT) is perceived as a cost-effective alternative to lessaffordable human interpreters, but little research has beendone on how people interact with such technology. Using aprototype translator application, we performed a formativeevaluation to elicit how people interact with the technologyand adapt their conversation style. We conducted two setsof studies with a total of 23 pairs (46 participants).Participants worked on storytelling tasks to simulate naturalconversations with 3 different interface settings. Ourfindings show that collocutors naturally adapt their style ofspeech production and comprehension to compensate forinadequacies in SLT. We conclude the paper with thedesign guidelines that emerged from the analysis.

Keywords

Multilingual communication, Spoken language translation, Automatic speech recognition, Machine translation

Discipline

Software Engineering

Research Areas

Software and Cyber-Physical Systems

Publication

CHI '15: Proceedings of the 33rd Annual ACM Conference on Human Factors in Computing Systems, Seoul, April 18-23

First Page

3473

Last Page

3482

ISBN

9781450331456

Identifier

10.1145/2702123.2702407

Publisher

ACM

City or Country

New York

Copyright Owner and License

Publisher

Additional URL

https://doi.org/10.1145/2702123.2702407

Share

COinS