Kansas Partnership for Accessible Technology

Brief Videoconferencing Captioning Overview

There are multiple forms of captioning that can be applied to videoconferencing, and options and implementation details vary according to the videoconferencing platform. It is important to understand their differences.

Live Captioning

The most effective form of captioning is that performed live by a professional captioner, typically through a service known as communication access real-time translation (CART). This should be the preferred solution for public meetings and other situations in which quality captioning is necessary and essential—and should always be available to persons with disabilities who request it. The trained human captioner can provide a number of advantages, including:

  • High accuracy
  • Awareness of names and domain-specific or technical vocabulary if materials are provided in advance
  • Ability to include information from (or cross-check interpretation of the spoken audio with) visual materials as well as contextual and situational cues
  • Ability to ask for clarification if necessary

Automated Captioning

Another form of captioning is automated captioning, ordinarily built into the videoconferencing software, in which speech-to-text technology generates the captions displayed. While the accuracy of such technology has improved greatly in recent years and continues to do so, it is still generally less than completely reliable. This form of captioning can be of some help, and is usually appropriate for small, informal conversations or other situations in which live captioning is not feasible.

Support in Commonly Used Platforms

Considering the videoconferencing platforms most commonly used by state agencies, at the time of this writing Microsoft Teams supports automated captioning (though it’s labeled “live captions” in the user interface), and Zoom supports (for at least some paid plans) both automated captioning and integration of live captioning through third-party CART. A general overview of the captioning capabilities of these and other platforms is available in this knowledge base.

Non-Embedded Captioning

While particular platforms may or may not support direct integration of captioning, particularly live captioning via third-party remote CART, essentially all can be used with non-embedded services that provide captions independently, typically in a separate browser window.

General Audio Considerations

All forms of captioning rely on high quality, clear, and audible audio for the captioner (whether human or computerized), so regardless of the type being used it is always important, as much as possible, for meeting organizers and speakers to:

  • Minimize background noise
  • Utilize networks of sufficient speed, bandwidth, and reliability to carry the audiovisual signal clearly, with minimal distortion, dropouts, etc.
  • Ensure each speaker has their own microphone
  • Speak directly into the microphone from a close distance
  • Speak clearly, and avoid speaking too quickly or quietly
  • Only speak one at a time

For more information, please see the other links provided in the Virtual Meetings and Remote Work section of the Kansas Partnership for Accessible Technology (KPAT) Resources page.