Teleconference Accessibility and Guidelines for Deaf and Hard of Hearing Users
Raja S. Kushalnagar, Christian Vogler · 2020 · Proceedings of the 22nd International ACM SIGACCESS Conference on Computers and Accessibility (ASSETS 2020) · doi:10.1145/3373625.3417299
Summary
This experience report from two Gallaudet University researchers examines the accessibility challenges that deaf and hard of hearing (DHH) users face in teleconference environments, drawing on nearly a decade of first-hand participation across platforms including Zoom, Microsoft Teams, Google Meet, Cisco WebEx, GoToMeeting, and Adobe Connect. Written with particular urgency during the COVID-19 pandemic — when Zoom usage grew over 500% in two months — the paper addresses a gap in accessibility guidance for remote meetings. The authors, both DHH themselves and User Interface and Experience experts, go well beyond simply recommending captions or sign language interpreters. They systematically catalogue the unique challenges of teleconferencing compared to face-to-face communication: the loss of social, emotional, and haptic feedback; the cognitive overload of juggling multiple visual elements simultaneously; the steep learning curves imposed by different platforms; and the hardware limitations that disproportionately affect DHH users. The paper distinguishes between two primary accommodation types — captions (both human-generated CART and automatic speech recognition) and sign language interpreters (via Video Remote Interpreting or Video Relay Services) — while noting that neither alone provides full functional equivalency. The authors then present detailed recommendations organized around user experience issues (screen complexity, remote troubleshooting, hardware diversity, backgrounds and lighting, audio quality, transcripts versus captions, eye contact, 2D versus 3D spatial awareness, and fatigue) and procedural guidelines (turn-taking, speaker identification, chat monitoring, and gallery view management).
Key findings
The paper identifies several critical insights from real-world teleconference experiences. DHH users must simultaneously manage interpreter video, speaker video, captions, presentation materials, chat boxes, and hand-raising tools — creating cognitive demands far exceeding those of hearing participants. Captions, while essential, fail to convey critical meta-speech information including speaker identification, sentiment, tone, and handling of overlapping speakers. Captioning accuracy drops significantly with background noise, complex terminology, or poor internet connections, and missing key words (nouns, identifiers) undermines satisfaction even when overall accuracy is high. Live text transcripts (CART) provide better functional equivalency than overlay captions because users can look away and catch up by reading back — matching how hearing participants can listen while briefly attending to other tasks. Wideband computer audio (3300-8000 Hz) significantly improves speech recognition and reduces mental effort for hearing aid and cochlear implant users compared to narrowband phone audio. Virtual backgrounds cause ghosting artifacts that are especially problematic for sign language, where arms, face, and hands move rapidly. The authors found that running a second instance of meeting software as a regular participant through a DHH lens helps organizers immediately identify problems invisible from the host view. Procedural changes — moderated turn-taking, mandatory speaker self-identification, designated chat monitors, and strategic use of video on/off — proved as important as technical features for achieving accessible meetings.
Relevance
This paper is highly relevant for any organization conducting remote or hybrid meetings with DHH participants, which became essentially every organization during and after the COVID-19 pandemic. The practical recommendations are immediately actionable: encourage headset microphones over laptop mics, provide step-by-step visual instructions before meetings, offer live transcripts alongside captions, implement moderated turn-taking, and designate a chat monitor. The insight that meeting procedures are as important as platform features challenges the common assumption that accessibility is primarily a technology problem. For accessibility practitioners, the paper provides a framework for evaluating teleconference platforms beyond simple checklist compliance — emphasizing functional equivalency and cognitive load parity rather than just the presence of captions or interpreter support. The legal landscape discussion around ADA applicability to teleconferences and the CVAA remains relevant as courts continue to address digital communication accessibility. The experience-based methodology, while not a controlled study, offers the kind of practical, lived-experience knowledge that complements empirical research.
Tags: deaf · hard of hearing · teleconferencing · remote participation · sign language · captions · video conferencing · cognitive load · communication access
Standards referenced: ADA · CVAA