当前位置: X-MOL 学术Vis. Comput. Ind. Biomed. Art › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Implementation and experiments for interactive lyrics transcreation system.
Visual Computing for Industry, Biomedicine, and Art ( IF 3.2 ) Pub Date : 2020-07-30 , DOI: 10.1186/s42492-020-00053-x
Ayano Nishimura 1 , Takayuki Itoh 1
Affiliation  

The evolution of the Internet has enabled us to enjoy the music created in various countries. But still, it is often difficult to understand the lyrics written in foreign languages. Professional translators have published many international songs with lyrics that fit the melody so that the ordinary people can enjoy the lyrics of such international songs. This paper discusses lyrics transcreation into the Japanese language. Also, the paper presents an interactive visual lyrics transcreation system and describes the details of its implementation. This system allows users to select temporary lyrics from a set of tentative translations and then freely modify the lyrics with a real-time visualization mechanism. We also propose a lyrics translation algorithm that solves an essential problem of lyrics translation into Japanese. In this study, we interviewed two experts regarding problems with lyrics translation and received reviews of our presented system. We also conducted preliminary experiments with 19 participants to determine the best combination of user interface components for our system. We performed additional user experiments inviting 12 participants to compare lyrics transcreation using the presented system to manual lyrics transcreation. Lyrics transcreation by the presented system brought better results against those of manual transcreation.

中文翻译:

交互式歌词转译系统的实现与实验。

互联网的发展使我们能够欣赏到各个国家创作的音乐。但是,仍然很难理解用外语写的歌词。专业翻译人员发表了许多带有适合旋律的歌词的国际歌曲,使普通百姓可以欣赏这些国际歌曲的歌词。本文讨论将歌词转换为日语的方法。此外,本文提出了一种交互式视觉歌词转译系统,并描述了其实现的细节。该系统允许用户从一组临时翻译中选择临时歌词,然后使用实时可视化机制自由修改歌词。我们还提出了一种歌词翻译算法,可以解决将歌词翻译成日语的一个基本问题。在这个研究中,我们采访了两位有关歌词翻译问题的专家,并收到了对我们提出的系统的评论。我们还与19名参与者进行了初步实验,以确定我们系统的用户界面组件的最佳组合。我们进行了另外的用户实验,邀请12位参与者将使用本系统的歌词转换与手动歌词转换进行比较。与手动翻译相比,本系统的歌词翻译带来了更好的效果。我们进行了另外的用户实验,邀请12位参与者将使用本系统的歌词转换与手动歌词转换进行比较。与手动翻译相比,本系统的歌词翻译带来了更好的效果。我们进行了另外的用户实验,邀请12位参与者将使用本系统的歌词转换与手动歌词转换进行比较。与手动翻译相比,本系统的歌词翻译带来了更好的效果。
更新日期:2020-07-30
down
wechat
bug