Research on a multilingual communication support environment that can appropriately convey the intentions of speakers

2021 Fiscal Year Final Research Report

• Project/Area Number: 18K18096
• Research Category: Grant-in-Aid for Early-Career Scientists
• Allocation Type: Multi-year Fund
• Review Section: Basic Section 61020:Human interface and interaction-related
• Research Institution: Osaka Institute of Technology
• Principal Investigator: Fukushima Taku 大阪工業大学, 情報科学部, 講師 (40714829)
• Project Period (FY): 2018-04-01 – 2022-03-31
• Keywords: 多言語間対話支援 / 用例対訳

Outline of Final Research Achievements

Machine translation technology has been developed, but complete machine translations are difficult to obtain at present. A parallel text that combines example sentences and their accurate translation is used in a multilingual communication support. In this study, we constructed a multilingual communication support environment that utilizes parallel texts and media-complex dialogue styles. The media-complex dialogue dialogue style includes non-textual information such as images and scales. In the experiments, we show that the proposed environment can accurately support in multilingual communication compared to conventional methods.

Free Research Field

HCI

Academic Significance and Societal Importance of the Research Achievements

メディア複合型対話様式を用いた多言語対話シート作成システムの実験より，従来は機械翻訳を利用せざるを得なかった質問への回答内容をより正確かつ短時間で質問者に伝達可能となる可能性を示した．また，穴あき用例を用いた多言語用例対訳作成手法の実験より，多言語話者の負担が少ない用例対訳作成が可能となること示した．本成果により，より正確な多言語間対話支援につながると考えられる．
ワードクラウドを活用した和菓子の味覚可視化手法の実験より，和菓子を食べる前後に存在していた味や食感の齟齬を減らすことができる可能性を示した．本成果により，より適切な文化の違いの伝達支援につながると考えられる．