Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Accented Character Entry Using Physical Keyboards in Virtual Reality

Snehanjali Kalamkar, Verena Biener, Daniel Pauls, Leon Lindlein, Morteza Izadifar, Per Ola Kristensson, Jens Grubert

TL;DR

The study tackles efficient accented-character entry on physical keyboards in VR by comparing three techniques: context-aware numeric codes (baseline), reconviguration, and multimodal eye-gaze-assisted input. In a within-subject user study (n=18), reconviguration and multimodal both delivered faster text-entry speeds than baseline, with notable gains for accented characters, and reconviguration also achieved higher usability. The findings show that VR enhancements—layout changes and gaze-assisted selection—can substantially improve multilingual typing efficiency, suggesting customizable, user-centered options for future VR keyboards. Overall, the work demonstrates practical paths to enable fluent, multilingual text entry in VR and informs design preferences for different user typists.

Abstract

Research on text entry in Virtual Reality (VR) has gained popularity but the efficient entry of accented characters, characters with diacritical marks, in VR remains underexplored. Entering accented characters is supported on most capacitive touch keyboards through a long press on a base character and a subsequent selection of the accented character. However, entering those characters on physical keyboards is still challenging, as they require a recall and an entry of respective numeric codes. To address this issue this paper investigates three techniques to support accented character entry on physical keyboards in VR. Specifically, we compare a context-aware numeric code technique that does not require users to recall a code, a key-press-only condition in which the accented characters are dynamically remapped to physical keys next to a base character, and a multimodal technique, in which eye gaze is used to select the accented version of a base character previously selected by key-press on the keyboard. The results from our user study (n=18) reveal that both the key-press-only and the multimodal technique outperform the baseline technique in terms of text entry speed.

Accented Character Entry Using Physical Keyboards in Virtual Reality

TL;DR

The study tackles efficient accented-character entry on physical keyboards in VR by comparing three techniques: context-aware numeric codes (baseline), reconviguration, and multimodal eye-gaze-assisted input. In a within-subject user study (n=18), reconviguration and multimodal both delivered faster text-entry speeds than baseline, with notable gains for accented characters, and reconviguration also achieved higher usability. The findings show that VR enhancements—layout changes and gaze-assisted selection—can substantially improve multilingual typing efficiency, suggesting customizable, user-centered options for future VR keyboards. Overall, the work demonstrates practical paths to enable fluent, multilingual text entry in VR and informs design preferences for different user typists.

Abstract

Research on text entry in Virtual Reality (VR) has gained popularity but the efficient entry of accented characters, characters with diacritical marks, in VR remains underexplored. Entering accented characters is supported on most capacitive touch keyboards through a long press on a base character and a subsequent selection of the accented character. However, entering those characters on physical keyboards is still challenging, as they require a recall and an entry of respective numeric codes. To address this issue this paper investigates three techniques to support accented character entry on physical keyboards in VR. Specifically, we compare a context-aware numeric code technique that does not require users to recall a code, a key-press-only condition in which the accented characters are dynamically remapped to physical keys next to a base character, and a multimodal technique, in which eye gaze is used to select the accented version of a base character previously selected by key-press on the keyboard. The results from our user study (n=18) reveal that both the key-press-only and the multimodal technique outperform the baseline technique in terms of text entry speed.
Paper Structure (29 sections, 3 figures, 4 tables)

This paper contains 29 sections, 3 figures, 4 tables.

Table of Contents

  1. Introduction
  2. Related Work
  3. Text Entry Techniques in VR
  4. Using Physical Keyboards in VR for Text Entry
  5. Typing Special Characters
  6. Multimodal Interaction Including Eye-Gaze
  7. Prototype Design
  8. baseline:
  9. reconviguration:
  10. multimodal:
  11. Implementation
  12. User Study
  13. Task
  14. Participants
  15. Procedure
  16. ...and 14 more sections

Figures (3)

  • Figure 1: Setup for the user study: Participant was wearing the Meta Quest Pro with the Logitech MX Keys S Keyboard in front of them.
  • Figure 2: Box plots for objective dependent variables: a) Overall WPM. b) Accented Character WPM. c) Regular Character WPM. d) Overall CER. e) Accented CER. f) Regular CER. BL=baseline, RVR=reconviguration, MM=multimodal. The number of stars indicates the significance levels between the techniques: *** < 0.001, ** < 0.01, * < 0.05.
  • Figure 3: Box plots for subjective measures: a) System Usability Scale Scores. b) NASA Task Load Index. c) Simulator Sickness. BL=baseline, RVR=reconviguration, MM=multimodal. The number of stars indicates the significance levels between the techniques: *** < 0.001, ** < 0.01, * < 0.05.