Entering text efficiently without a traditional keyboard is a widespread challenge in the XR headset industry. Researchers have developed a vast catalog of existing text entry methods, assessing their advantages and drawbacks. By making this resource accessible, the team hopes to inspire further innovation in creating new and improved techniques.
In a guest article by Max Di Luca, Associate Professor at the University of Birmingham and head of the VR Lab there, he shares his insights. Di Luca, who previously worked at Meta advancing VR hand inputs and haptics, was recently honored by the ACM SIGCHI 2025 awards for his work in developing foundational interaction frameworks for Android XR through strong industry and academic partnerships.
As immersive experiences like VR and AR grow more complex, efficient text entry becomes more critical for smooth interactions. Whether it’s drafting emails, logging in, or socializing in the metaverse, seamless text input is vital for the functionality of XR applications.
To tackle this issue, my team at the VR Lab collaborated with global institutions, including the University of Copenhagen, Arizona State University, the Max Planck Institute for Intelligent Systems, Northwestern University, and Google, to launch the XR TEXT Trove. This initiative catalogs over 170 text entry techniques specifically for XR, serving as a comprehensive guide that highlights the strengths and weaknesses of various methods developed academically and in the industry.
The techniques in the trove are organized with 32 different codes, covering 13 interaction attributes such as input device and body part used for input, as well as 14 performance metrics like words per minute and error rate. This extensive catalog provides a broad view of the state of XR text entry.
One key finding from our research is that text input performance is closely tied to the number of input elements, like fingers or controllers. Multi-finger typing can achieve speeds similar to traditional keyboard typing, with each additional finger adding approximately five words per minute, as shown in our data visualizations.
Our study also finds that haptic feedback, using external surfaces, and focusing on fingertip visualization can significantly enhance typing efficiency. Typing on a surface rather than in mid-air, for example, can reduce muscle strain and improve comfort, avoiding issues like Gorilla Arm Syndrome.
Interestingly, despite many alternatives, the keyboard format remains unmatched for high-speed typing, possibly due to its steep learning curve. We foresee advancements in speed by reducing travel distances on multi-finger keyboards through machine learning and AI, akin to the ‘swipe typing’ revolution on smartphones.
The XR TEXT Trove project is a notable advancement towards better understanding and improving text input in virtual and augmented reality. By providing a well-organized and searchable database, we aim to aid researchers and developers in crafting efficient text entry solutions for immersive experiences.
Our full findings will be shared at the upcoming ACM CHI conference in Yokohama, Japan. Similar to our previous work with the Locomotion Vault, which catalogs VR locomotion techniques, the XR TEXT Trove is designed to assist researchers and designers in refining and advancing XR text input methods.
