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

Text input without a traditional keyboard is a persistent hurdle in the extended reality (XR) landscape, challenging productivity potential. In an exciting move, researchers have systematically compiled existing text entry methodologies, weighing their strengths and weaknesses. This project aims to empower innovators to design better approaches by offering this information openly.

Meet Massimiliano Di Luca, who directs the VR Lab at the University of Birmingham and has forged significant advancements in XR user interactions during his tenure at Meta. His collaborative efforts have earned accolades, notably advancing Android XR’s interaction framework through effective industry-academia partnerships.

The ever-growing sophistication of virtual and augmented reality (VR/AR) experiences spotlights the need for efficient text input, foundational to engaging with digital worlds—whether it’s for work emails in a virtual office or socializing in the metaverse. The efficiency of text entry determines the overall usability of XR platforms.

To tackle this, our diverse team from respected institutions across the globe, including the University of Birmingham, collaborated to create the XR TEXT Trove. This resource meticulously catalogs over 170 text input techniques for XR environments. The database serves as both a repository and a filtering tool to evaluate each method’s merits and drawbacks.

Organized through 32 distinct codes, the techniques are assessed based on interaction attributes—like the type of input device and body part used—and performance metrics such as words per minute (WPM) and total error rate (TER). This comprehensive analysis offers an in-depth view of the current state of text entry in XR.

One key finding from our research highlights that the speed of text input is largely constrained by the number of input elements—the more fingers or control points involved, the faster the typing. Multi-finger typing is the only method that competes with traditional keyboard touch-typing speeds, accelerating with each additional input element.

Furthermore, incorporating haptic feedback, typing on external surfaces, and relying on fingertip visual feedback enhances the typing experience. These techniques reduce strain and improve comfort, effectively minimizing issues like Gorilla Arm Syndrome.

Interestingly, despite innovations, no alternative surpasses the traditional keyboard’s efficiency in WPM, likely due to the steep learning curve associated with newer methods. Advancements in machine learning could potentially revolutionize XR typing by shortening key travel distances on multi-finger inputs, echoing the efficiency gains smartphones saw with swipe typing.

The XR Text Trove initiative represents a pivotal step towards understanding and evolving text input in immersive realities. By creating an accessible, structured database, we empower researchers and developers to innovate more user-friendly solutions for the future.

The insights we share in our published paper promise substantial benefits to the XR community. Our database and corresponding tools are accessible on the XR TEXT Trove website. Eager to share our work with a wider audience, we’ll present at the prestigious ACM CHI conference in Yokohama, Japan, next month.

Our team, including contributors to the Locomotion Vault, is committed to advancing XR research and design, hoping to give the community a running start towards improved interaction methods.