To support of Michigan Tech’s Accessible Information and Communication Technology Policy 1.15, all video content posted on websites or in the Canvas learning management system (LMS) should include accurate captions to ensure access for all users, as required under WCAG 2.0 success criteria 1.2.2 (Level A):
Captions (Prerecorded): Captions are provided for all pre-recorded audio content in synchronized media, except when the media is a media alternative for text and is clearly labeled as such.
Captions Benefit Everyone
Captions on videos support viewers with a hearing impairment. The captions display information from the audio track of the video, including speaker identity and certain non-speech information (such as sound effects) that contain important information.
Consider the benefits of captions for other users as well:
- Support for non-native speakers
- Video that contains complex or unfamiliar jargon and vocabulary
- Supports improved learner comprehension
- Speaker has accented English
- Audio is too low or contains background noise or other distractions
- Provides searchable metadata for users to efficiently locate specific information
Types of Captions
Captions can be either open and closed. Open captions are “burned” into the video image and cannot be turned off or modified in any way. Closed captions can be turned on or off from within the video player. Both open and closed captions use the same audio transcript file that is then synchronized to the video timeline. For our purposes we will be reviewing the creation and use of closed captions.
Captions can be manually created or machine generated using automatic speech recognition (ASR) technology. Although ASR captions are becoming more accurate they are not considered fully compliant due to inherent errors with word selection and formatting. Auto-generated captions are a great starting point, but should be reviewed for errors by the creator or other subject expert. Correcting these errors will ensure accuracy and avoid potential confusion for users.
Panopto-Huskycast (Michigan Tech’s video platform) generates ASR captions which can be inserted into videos from the available online editor. This editor is also where creators can correct ASR caption errors to ensure full accuracy. Recordings made in Michigan Tech’s Zoom account can also leverage ASR audio transcripts for cloud-based recordings. Other hosting platforms, like YouTube, offer ASR captions and also have options to make manual corrections in the Video Suite.
Huskycast, is Michigan Tech’s branded Panopto video platform that is used by instructors to host video content for use in their courses. Huskycast is also used for non-instructional uses on campus. Michigan Tech’s Canvas learning management system integrates closely with Huskycast to make uploading and sharing videos with students an easier process.
Inserting ASR Captions
Once content is uploaded to Huskycast content owners (called creators in Huskycast) can follow the steps below to add ASR captions to their videos and then make corrections for accuracy:
- Open the Panopto online editor for the session you wish to add captions to by choosing the Edit option.
- Select the Captions option from the left side of the editor and click the Import Captions drop-down menu.
- If no captions exist for the session choose the Import automatic captions option to insert the ASR captions generated by Panopto.
- Captions should now appear with the associated time that they appear in the video
Editing ASR Captions
Huskycast video creators should review the ASR captions for accuracy and make any corrections necessary following the steps below:
- Review ASR captions in the editor and identify any errors
- Click on the three dots next to a caption to access the Edit mode
- Make the necessary corrections
- You can also choose the Delete option to remove the caption entry
- Click the Apply button at the top right corner of the editor to commit all changes to the session for processing.
Live Transcription (Captions)
Michigan Tech Zoom accounts feature a live captioning option for cloud-based recordings. When enabled by the meeting host, attendees can view these captions at the bottom of the meeting screen, or as a full transcript window in a sidebar. Zoom hosts can also allow attendees to save the transcript file if desired. To use this feature you will need to enable the live transcription feature in your Zoom account advanced meeting settings. Refer to the resources below for more information about using this feature.
An automatic transcription feature is available for Zoom cloud recordings. This feature creates an audio transcript file of meeting audio after the meeting has ended. Once complete the audio transcript is available from the cloud recordings area of the host’s Zoom account. An available integration with Panopto-Huskycast is also available (see below) which can automatically transfer the audio transcript file and the meeting recording to Huskycast.
A Zoom integration is available in Michigan Tech’s Panopto-Huskycast platform that allows Zoom users to designate a Huskycast folder where Zoom cloud recordings (created with a recurring Meeting ID) can automatically be transferred to. Zoom audio transcripts generated with these cloud recordings will also transfer to Huskycast and be available to viewers from the “CC” button in the player controls. Review the knowledge base article for more information.
- Zoom live transcription feature
- Using audio transcription with Zoom cloud recordings
- Using the Panopto-Huskycast-Zoom integration
- Video captioning options for YouTube (University Marketing and Communications resource)