Virginia Tech® home

Captioning and Transcripts

Insert your title here

Captioning and Transcripts

Captioning

Captioning video content is an essential part of creating high-quality educational resources that improve student outcomes and user engagement. Captions and transcripts provide a text alternative to audio, offering viewers additional ways to access content, improving overall comprehension, and supporting all audiences. Captions improve comprehension by removing common miscommunications stemming from variations in:

  • hearing ability

  • speaker accent

  • audio quality

  • listener’s primary language

  • complexity of the subject matter

  • missed audio content

Terminology

Captioning terminology can be confusing. Establishing a shared understanding of the terms used is important, for example: 

  • Captions are synchronized text of spoken dialogue, sound effects, or other audio cues that appear within a video. Closed captions can be turned on and off, open captions are locked on.

  • Transcripts document the content from an audio recording (e.g., spoken dialogue, sound effects, or other audio cues) within a separate text file not time-synchronized to the video. Interactive transcripts are time-synchronized text versions of audio. They are typically searchable, and user selection of the text jumps the video to that specific location.

  • Captions and transcripts can be machine-generated (automated speech recognition or ASR), or generated by a human captionist, or both. 

  • Enhanced automated speech recognition (ASR) captions use preprogrammed vocabulary and other preparatory materials to maintain an accuracy rate of at least 95%.

  • Live captions are generated in real time as audio is captured during a live event, while post-production captions are generated for previously recorded content. 

Live Captioning

Live captioning transmits speech and audio into real-time text by generating text that corresponds to the auditory information. Live captions may come from a machine, a human, or a combination of both. When live captions are available (e.g., in Zoom, or Panopto), users can enable this feature to display all spoken content as live text.

Live captioning is required for any live, digital, public-facing event across the university (e.g., webinars, seminars, and town halls). Synchronous live captions can be integrated into Zoom or Microsoft Teams or provided via a separate browser window for the end user. As outlined below, Live Captioning is a requestable service for university faculty and staff. 

Live Event Caption Planning

When planning a live event, it is important to plan for live captioning. Centralized funding covers the cost of CaptivateTM Automated Speech Recognition (ASR) captions for live events, at no cost to the event organizer. These captions are about 95% accurate without preparatory materials. Other options for live captions include Zoom or Microsoft Teams ASR captions. Please refer to the chart below to determine the best option for your live event.

 

FunctionalityCaptivate ASRZoomMicrosoft Teams
UtilizationSchedule at least two business days in advanceTurn on at the start of the meetingTurn on at the start of the meeting
Accuracy

Higher captioning accuracy

  • About 95% accurate without preparation materials

  • Between 95% – 99%+ accuracy with preparation materials.

Lower captioning accuracy

Lower captioning accuracy

  • Microsoft does not provide accuracy ratings at this time

Non-speech audio eventsIncluded (e.g., door closing, dog barking, etc.)Not includedNot included
Speaker identificationIncludedIncluded

Included in default

  • Individual users can choose not to be identified

Appearance Settings

Livestream link allows the user to change font type, size, and color contrast

  • Users can view captions on a separate device and/or in a separate window, as well as in Zoom and/or Microsoft Teams

Option to change font type, size, and color contrast

  • Users can only view captions in Zoom meetings

Option to change font type, size, and color contrast

  • Users can only view captions in Microsoft Teams meetings

SpeedReal-time, in-meetingReal-time, in-meetingReal-time, in-meeting
CostNo cost to VT userNo cost to VT userNo cost to VT user
Post-event TranscriptPost-event transcript providedDownload the transcript prior to ending the Zoom meetingDownload the transcript from the meeting event calendar

 

If using CaptivateTM captions, please submit a Live Captions Request form. When possible, event planners can increase captioning accuracy for their event by providing event preparation materials at least two business days in advance. Examples of preparation materials include:

  • Technical vocabulary

  • List of speaker names

  • Speaker biographies

  • Presentation materials (e.g., slide deck)

  • Event program or script 

For support in determining the best live captioning plan for an event, please email captioning@vt.edu

Post-Production Captioning

For pre-recorded videos, such as news announcements or course lectures, disability legislation requires post-production captions. For all videos uploaded to Panopto, machine captions are generated using automatic speech recognition (ASR) software. Accuracy is typically around 95%, though factors such as audio quality, the speaker, and specialized or complex terminology may impact accuracy. Any recordings shared through a public-facing website or Canvas course should be edited for accuracy using Panopto’s built-in captioning editor. Please refer to the Captioning in the Panopto 4Help article for more information.

Transcripts

Transcripts are necessary for audio-only content, such as podcasts. Audio and video files in Panopto include an interactive transcript that viewers can access by selecting “Captions” on the player’s menu bar. When using Live Captioning services integrated with Zoom or Microsoft Teams, a live transcript is also available. Users can order transcripts for audio recordings outside of Panopto by submitting the Transcription Request form

Questions

Please email captioning@vt.edu with questions about captioning or transcripts.