# Transcripts are tied to eachother

**URL:** <https://discuss.openedx.org/t/transcripts-are-tied-to-eachother/7745>\
**Category:** Educators\
**Created:** [July 5, 2022, 2:37pm UTC](https://discuss.openedx.org/t/transcripts-are-tied-to-eachother/7745 "2022-07-05T14:37:16Z")\
**Posts on this page:** 3\
**Page:** 1

<div class="post-metadata">

**Author:** ![mbaileyexasol](https://avatars.discourse-cdn.com/v4/letter/m/9f8e36/32.png) [@mbaileyexasol](https://discuss.openedx.org/u/mbaileyexasol)\
**Post date:** [July 5, 2022, 2:37pm UTC](https://discuss.openedx.org/t/transcripts-are-tied-to-eachother/7745/1 "2022-07-05T14:37:16Z")

</div>

Let me know if you’ve seen this one before:

We built a new course, let’s call it B. We then exported the contents of course A, and imported it into course B. We can change anything we want with B, but when we insert a new transcription into the B video, the A course’s video that B was copied from has it’s transcript updated as well.

To make this clearer, it has nothing to do with the video. In the B course I linked to a completely separate youtube video, then added the transcript in edx. The transcript for the new video synced back into course A.

Has anyone else experienced this? It’s like when we created the new course and imported from the old one, the system created all new object\_ids for each individual component, except for the transcripts.

---

<div class="post-metadata">

**Author:** ![Rohan](https://avatars.discourse-cdn.com/v4/letter/r/f19dbf/32.png) [@Rohan](https://discuss.openedx.org/u/Rohan)\
**Post date:** [July 14, 2022, 12:48pm UTC](https://discuss.openedx.org/t/transcripts-are-tied-to-eachother/7745/2 "2022-07-14T12:48:33Z")

</div>

I’ve definitely seen that. When exporting and re-importing courses to the same site, AFAICT, new ID’s _aren’t_ actually generated. It’s on my todo list to eventually create a script to force-randomize IDs before import, so that I can store exported classes + transcripts into a git repository and then update transcripts in the repo, and then re-randomize and re-import (instead of needing to manually add every transcript over and over again as they change.)

---

<div class="post-metadata">

**Author:** ![TonyH](https://sea2.discourse-cdn.com/flex020/user_avatar/discuss.openedx.org/tonyh/32/4079_2.png) [@TonyH](https://discuss.openedx.org/u/TonyH)\
**Post date:** [August 5, 2022, 3:33pm UTC](https://discuss.openedx.org/t/transcripts-are-tied-to-eachother/7745/3 "2022-08-05T15:33:09Z")

</div>

It would be fantastic if this issue would be fixed! We have seen this problem too frequently among our course developers.

- We saw this early on when we had a _template course_ with an embedded video. After that, we reindexed all IDs before importing the template to solve the problem.
- Now, the most common way it occurs is when a developer duplicates a section/subsection/unit with a video that has subtitles.

This bug **causes a significant problem** because it is futile to try to remove the wrongly-linked subtitles. Unfortunately, the subtitles for other videos can show up for videos that do not have a transcript linked to them on YouTube, so there is no way to replace or overwrite them.

- In some instances, the only way to solve the problem is to download the video from YouTube and upload it to the developer’s personal YouTube account to generate a new ID (if the video is not flagged for copyright violation).
- Another workaround we have used is to upload an empty subtitle file and then turn off the subtitle for the video in the video component. It is not ideal because the viewer can turn on the subtitles, but an empty side panel is better than displaying the incorrect text.