Skip to Main Content

Transcripts for Audio and Video in Collection Loader

  • Updated

On this page:

When uploading audio or video files to Collection Loader, transcripts are automatically generated for content containing spoken word or singing in supported languages. (See a full list of supported languages and language-specific features. Please note that transcription accuracy is subject to the quality and clarity of the audio provided.)

Along with the item's metadata, machine-generated transcripts are also searchable on JSTOR, allowing the content of your audio and video contributions to appear in search results.

Transcript with clickable timestamps alongside a video media player on JSTOR

When viewing an audio or video item on JSTOR, users can view the transcript alongside the media and click on the embedded timestamps to jump to specific points in the recording. For videos, transcripts are also used for closed captions in the media player.

Uploading transcripts

Currently, the Video Text Tracks (VTT) file format is supported for transcripts. If you download an item's machine-generated transcript, it will be in the VTT (.vtt) format.

To display correctly on JSTOR, uploaded transcripts must open with the string WEBVTT and cue timestamps must be specified in the format: hh:mm:ss.ttt (hours, minutes, seconds, milliseconds).

Below is an example of a VTT file with cue identifiers and timestamps:

WEBVTT

0
00:00:00.009 --> 00:00:02.359
The Pacific Sea Nettle, or West Coast Sea Nettle,

1
00:00:02.369 --> 00:00:05.199
can be found in the Northeastern Pacific Ocean,

2
00:00:05.429 --> 00:00:08.180
in temperate to cooler water off British Columbia,

3
00:00:08.369 --> 00:00:11.949
and the West Coast of the United States ranging south to Mexico.
Note: The VTT file format supports enhanced text formatting, styles, and comments; however, these styles won't be displayed in Collection Loader or when published to JSTOR. In addition, any HTML markup in the transcript content will be rendered as plain text.

Transcript tab and Replace transcript menu for an item in Collection Loader

To upload a transcript for an audio or video item:

  1. Log into the Collection Loader and choose the Collections link.
  2. On the Collections page, choose your collection and then the item you wish to edit.
  3. Select the Transcript tab to view the item's current transcript.
  4. Open the Replace transcript drop-down menu and select Upload new transcript.
  5. Click the Choose file button, and select your VTT (.vtt) file.
  6. Click the Upload button.

Saving and re-publishing the item is not required when uploading a new transcript. If you're not seeing your updated transcript on the item page on JSTOR, you may need to clear your browser cache and refresh the page.

You can restore the original machine-generated transcript for an item at any time by choosing Reset transcript in the Replace transcript menu.

Note: At this time, only the initial machine-generated transcript for an item is searchable on JSTOR. User-uploaded transcripts will be displayed on the item page and in closed captions, but the content won't be included in searches.

Viewing and downloading transcripts

You can download an item's transcript, whether it's the original machine-generated transcript or user-provided, as a Video Text Tracks (VTT) file.

To download a transcript for an audio or video item:

  1. Log into the Collection Loader and choose the Collections link.
  2. On the Collections page, choose your collection and then the item you wish to view.
  3. Select the Transcript tab to view the item's current transcript.
  4. Click the Download transcript button to export the current transcript as a VTT (.vtt) file.

You can then edit the downloaded file in your editor of choice (such as a text editor) and re-upload it to Collection Loader, or provide your own transcript file.

Articles in this section