Scale Up Multilingualism in Health Emergency Learning: Developing an Automated Transcription and Translation Tool

Heini Utunen, Thomas Staubitz, Richelle George, Yu Ursula Zhao, Sebastian Serth, Anna Tokar

Research output: Contribution to journalArticleScientificpeer-review

2 Citations (Scopus)
6 Downloads (Pure)

Abstract

World Health Organization's (WHO) emergency learning platform OpenWHO provided by Hasso Plattner Institut (HPI) delivered online learning in real-time and in multiple languages during the COVID-19 pandemic. The challenge was to move from manual transcription and translation to automated to increase the speed and quantity of materials and languages available. TransPipe tool was introduced to facilitate this task. We describe the TransPipe development, analyze its functioning and report key results achieved. TransPipe successfully connects existing services and provides a suitable workflow to create and maintain video subtitles in different languages. By the end of 2022, the tool transcribed nearly 4,700 minutes of video content and translated 1,050,700 characters of video subtitles. Automated transcription and translation have enormous potential as a public health learning tool, allowing the near-simultaneous availability of video subtitles on OpenWHO in many languages, thus improving the usability of the learning materials in multiple languages for wider audiences.

Original languageEnglish
Pages (from-to)408-412
Number of pages5
JournalStudies in health technology and informatics
Volume302
DOIs
Publication statusPublished - 18 May 2023
Publication typeA1 Journal article-refereed

Keywords

  • Humans
  • Multilingualism
  • COVID-19
  • Pandemics
  • Language
  • Translating

Publication forum classification

  • Publication forum level 1

Fingerprint

Dive into the research topics of 'Scale Up Multilingualism in Health Emergency Learning: Developing an Automated Transcription and Translation Tool'. Together they form a unique fingerprint.

Cite this