Chat Translation Error Detection for Assisting Cross-lingual Communications

TL;DR

This paper presents a system for detecting translation errors in chat conversations to improve cross-lingual communication, utilizing a new bilingual chat corpus and an error detection model.

Contribution

It introduces a novel Japanese-English chat corpus and a baseline error detection system for improving machine chat translation accuracy.

Findings

Constructed BPersona-chat corpus with crowdsourced quality ratings

Developed a baseline error detector for chat translation errors

Provides a foundation for advanced error detection systems

Abstract

In this paper, we describe the development of a communication support system that detects erroneous translations to facilitate crosslingual communications due to the limitations of current machine chat translation methods. We trained an error detector as the baseline of the system and constructed a new Japanese-English bilingual chat corpus, BPersona-chat, which comprises multiturn colloquial chats augmented with crowdsourced quality ratings. The error detector can serve as an encouraging foundation for more advanced erroneous translation detection systems.