Researchers Compile Over 170 Text Input Techniques to Enhance Typing in XR

The quest for efficient text entry in XR headsets, devoid of a physical keyboard, remains an open puzzle. Researchers have put together a detailed collection of existing text input techniques, aiming to provide a comprehensive set of methods and a balanced view on their respective strengths and weaknesses. By openly sharing this catalog, they intend to spur the development of new and more effective techniques.

Massimiliano Di Luca, who heads the VR Lab at the University of Birmingham, UK, brings a wealth of experience from his stint at Meta, where he paved the way in hand input and haptics for VR. His collaboration with industry players on the Android XR interaction framework won accolades at the ACM SIGCHI 2025 awards, underlining the potential for combining academic insight with industry needs.

As we gravitate toward more sophisticated immersive experiences, effective text input is critical for seamless interactions in virtual and augmented realities. Whether drafting emails in a virtual office or logging into social platforms in the metaverse, having quick and intuitive text input is indispensable for XR application usability.

To tackle this critical issue, our team at the VR Lab developed the XR TEXT Trove, in collaboration with researchers from various prestigious institutions and Google. This initiative aims to categorize and examine over 170 text entry methods specific to XR environments. The TEXT Trove acts as a structured and searchable resource that highlights the benefits and limitations of the different text input technologies developed so far.

Our catalog employs 32 different parameters to classify techniques—13 of which focus on interaction attributes like input devices and body parts used, while 14 are centered on performance metrics like typing speed and error rates. The rich array of techniques and their detailed categorization offer an insightful snapshot of the progress in XR text entry methods.

Through our research, we’ve discovered that the speed of text input is inherently restricted by the number of input elements—be it fingers, controllers, or any other tools. Users who type with multiple fingers can potentially match the typing speed seen with traditional keyboards. As shown in the accompanying data plots, each added input element can amplify typing speed by approximately 5 words per minute.

Additionally, factors like haptic feedback and the use of external surfaces significantly enhance typing efficiency. Typing against a surface, rather than in mid-air, not only boosts comfort but may also reduce muscle strain, alleviating issues commonly known as Gorilla Arm Syndrome.

Despite advancements, the conventional keyboard still reigns supreme in terms of typing speed, possibly due to its steep learning curve. Our speculation is that achieving faster typing in VR than on PCs might come from innovations using Machine Learning and AI to reduce finger travel distances on a keyboard. XR environments need an adaptation akin to ‘swipe typing’ on smartphones, which transformed single-finger typing.

Thus, the XR Text Trove is a significant contribution toward better understanding and improving text input in virtual and augmented realities. We have made this database accessible to bolster further research and development, aiming for more efficient, user-friendly text entry solutions in the immersive landscapes of the future. Our findings will be showcased in an upcoming presentation at the ACM CHI conference in Yokohama, Japan.

Several members of our team were also instrumental in the creation of the Locomotion Vault, a similar initiative aimed at cataloging VR movement techniques, thereby offering insight and a head-start for continued improvement in XR development.