Work

Creating an interactive training environment for dialogue interpreting

Year: 2025
Type: article
Abstract: Community and public service interpreters (CPSI) interpret encounters in health care, legal or social settings. The skills required to interpret in these settings are similar to those of conference interpreters (Tiselius, 2021). Compared to conference interpreters, however, many CPSI seem to receive only short training (Hanft-Robert & Mösko, 2024). In a recent project, we explored the feasibility of creating a virtual reality app (VR) that leverages AI to generate bilingual, interactive dialogues for dialogue interpreting practice. VR has already been tested as a means to enhance the training of CPSI. They work with pre-recorded dialogues (Chan, 2022; Eser et al., 2020; Gerber et al., 2021) or video chat with other trainees (Braun et al., 2013, 2020). The reports published on these applications show that students experience these applications usually as very engaging and relevant for their training (Braun et al., 2020; Chan, 2022; Eser et al., 2020). But pre-recorded dialogues, though authentic, have important limitations as the trainee has no possibility to interact with the VR avatars. Video-chats offer the opportunity for live role-play and hence, more interaction, but require co-trainees who share the working language. To overcome these limitations, we tested the use AI to generate a bilingual dialogue. The dialogue was embedded in a VR environment that simulated an encounter between a patient and a doctor. The aim was to generate dialogues that 1) dynamically adapt to the interpretation, 2) resemble encounters typically mediated by CPSI, 3) use typical migrant languages which such as Tigrinya, Dari or Kurmanji. A prototype of the app was tested in a user observation with 11 CPSI to assess the practicality and feasibility of this idea. Each participants tested the app with one of the 20 most common migrant languages in Switzerland, filled in the User Experience Questionnaire (Laugwitz et al., 2008) and SUS (Brooke, 1996), and described their experience in a post-task interview. We will present the functionalities of the prototype and report on the results user observation.
Language: English
Open Access status: green
OpenAlex ID: W7120875895
FWCI:
Cites:
Cited by:
Related to: 9