Overview

Kaltura users have the ability to upload a text-only transcript to an entry and have the system generate a time-based caption file that will display while the video plays. 

While Kaltura users can edit their captions either by using Kaltura's web-based editor or by downloading the captions and editing them on their computer, they can also choose to work with a text-only version (without any timecodes) use Kaltura's "machine alignment" feature to have it generate timings for a new caption file, which will display the words on screen at the appropriate time while the video is being played. 

This feature is available to all Kaltura users at UC San Diego (faculty, students, staff, etc.). A transcript can be uploaded in either Canvas or MediaSpace. 

Critical Concepts

What's the Difference between Captions and Transcripts?

Sometimes the terms "captions" and "transcripts" are used interchangeably, but there are meaningful differences, particularly within the context of media management. While both contain a textual representation of words spoken during a video, here's where they diverge:

• Caption files contain timecodes that indicate the start and endpoints of when a piece of text should be on screen while the video is playing. They can have a variety of file extensions, such as .srt or .vtt.
• Transcripts contain no timecodes at all, and always have a .txt file extension. It's just a big block of text.

Kaltura generates its transcripts based on the captions. So if you edit the captions, the changes will be reflected in the transcript. The transcript can be displayed within the media player when viewing a video in My Media, a Media Gallery, or a MediaSpace channel. Most embedded videos provide the transcript as well. One nice thing about Kaltura's transcript  is that it will highlight the text that's currently being spoken, and you can click on parts of it to jump to that part in the video. 

Things to Keep in Mind

• Why edit a transcript and not captions? As you can see from the image above, caption files contain timecodes as part of their text. If you're editing the text while listening to it, it may be easier for you to deal with just the words without any timecodes in the way. In addition, you have the ability to add paragraph breaks, which makes the overall text more legible. This could come in handy if you elect to generate a text-based version of your video content at some point (a best practice embraced by the Universal Design for Learning instructional design framework).
• We discuss how to download the transcript that Kaltura generates elsewhere. If you want to download and edit a transcript that Kaltura has created, you can follow our instructions in another knowledge base article:
  • How to download a transcript for a Kaltura entry
• You must have a .txt file. In order to upload a file as a transcript, it must be a file with a .txt extension. You cannot upload a word document, a PDF, or any other type of document. 

Steps to Take

Uploading a transcript can be done in either Canvas or MediaSpace. Follow steps 1 and 2 depending on where you want to perform the action.

3. Find the entry in question and click on its title or thumbnail.
4. Under the media player, click Actions and select + Caption & Enrich from the menu that appears.
5. In the "Order Captions & Enrichment Services" section that populates below, ensure that the following options are selected:
   1. Service: Machine
   2. Source Media Language: English
   3. Feature: Alignment
6. Click Upload txt.
7. Click + Select File.
8. Locate and select the transcript on your computer and click Open.
9. Enter a title and description for the transcript if desired. (These are optional.) They'll be visible if you or viewers click on the "attachments" tab under the media player.
10. Click Save.
11. Select the radio button for the text file you just uploaded.
12. Click Submit.

You should see a message appear indicating that your request has been received.

You'll receive an email notification once the machine alignment process is complete. The process will create a new set of captions that will become the default captions for the entry. It will also hide any machine captions from viewers that were previously visible in the player.