Video Conferencing Tools: Comparative Study of the Experiences of Screen Reader Users and the Development of More Inclusive Design Guidelines
Barbara Leporini, Marina Buzzi, Marion Hersh · 2023 · ACM Transactions on Accessible Computing · doi:10.1145/3573012
Summary
This comprehensive study investigates the accessibility and usability of three major video conferencing platforms—Zoom, Google Meet, and Microsoft Teams—for blind and visually impaired people using screen readers. Conducted during the COVID-19 pandemic when video conferencing became essential for employment, education, and social participation, the research addresses a critical gap in understanding how well these tools serve screen reader users. The methodology combined expert inspection evaluation with end-user surveys. Two accessibility experts (one blind since childhood) conducted cognitive walkthroughs of nine key participant functions: joining meetings, raising hands, controlling microphone/camera, screen sharing, using chat, accessing shared content, checking device status, and viewing participant information. Separate evaluations covered desktop versions (using JAWS with Windows) and mobile versions (using VoiceOver on iOS). End-user surveys collected responses from 65 blind participants for desktop tools and 94 for mobile platforms, gathering both quantitative ratings and qualitative comments about their experiences. The study evaluated each function across three dimensions: keyboard/gesture focus support, availability of shortcuts, and quality of screen reader feedback. This granular approach revealed that while all three tools technically support basic functions, the practical experience varies dramatically based on interface complexity, shortcut availability, and clarity of audio feedback.
Key findings
Zoom emerged as the most accessible option on desktop, with 81.5% of users satisfied or very satisfied, compared to 56.7% for Teams and 37.5% for Meet. On mobile devices, Zoom maintained its lead (73.8% satisfaction) though Google Meet performed relatively better in the mobile environment. However, none of the three tools achieved full accessibility—all had significant barriers for screen reader users. Basic functions like joining meetings and toggling microphone/camera were generally accessible across all platforms. However, more complex tasks proved problematic: only 30% of Zoom desktop users could identify who had raised their hand, only 12% of Meet users could check other participants' camera status, and accessing shared content was largely inaccessible on mobile (rendered as images without text alternatives). Keyboard shortcuts were crucial for desktop usability—96% of Zoom users employed shortcuts sometimes or always, compared to only 70% for Meet and 73% for Teams. The chat function presented significant challenges: while the chat panel could be reached, reading message history required multiple Tab and Shift+Tab presses, and copying links or text from chat was difficult. Users reported cognitive overload from competing audio streams—meeting speech versus screen reader output—making multitasking extremely demanding. Participant comments highlighted interface complexity ("these tools are hellish") and the lack of easily available shortcuts. Six participants reported needing sighted assistance to learn the interfaces initially, particularly with Teams.
Relevance
This research provides actionable guidance for video conferencing developers through six detailed guidelines (G1-G6) that go beyond existing W3C-WAI recommendations. Key recommendations include: organizing interfaces into logical functional panels accessible via shortcuts; providing clear, unambiguous screen reader feedback (e.g., "mic on" not "turn on"); offering audio assistance for visual tasks like camera framing; and enabling customization of notifications to reduce cognitive overload. For accessibility practitioners, the study demonstrates that WCAG compliance alone does not guarantee usability for screen reader users. Functions may be technically accessible but require so many steps or cause such cognitive strain that they become impractical. The finding that desktop applications were preferred over web versions (86% for Zoom, 75% for Meet) suggests organizations should ensure native app options are available when screen reader users are expected. The research methodology—combining expert inspection with user surveys—offers a model for evaluating complex application accessibility. The specific function-by-function analysis provides a template for testing similar collaborative tools. As remote work becomes permanent for many, these findings highlight the need for continuous accessibility improvement in essential communication infrastructure.
Tags: video conferencing · screen readers · blind users · Zoom · Microsoft Teams · Google Meet · remote work · COVID-19 · usability
Standards referenced: WCAG 2.1 · EN 301 549 · Section 508 · VPAT