Bazarr embedded sub extractor removing words

[anonyme123]

Describe the bug Hello, it seems that Bazarr's embedded subs extractor is removing some words. We can see that it extracted the embedded English subs:

But it removed the word "SPAN" from the sub file:

I re-downloaded the original file outside of Bazarr and the word was not removed:

Here are my Post-Processing settings, I don't know if they are also used during embedded subs extraction:

To Reproduce Steps to reproduce the behavior:

1. Configure Bazarr to extract embedded subs
2. Download a video that has embedded subs with the word 'SPAN' in them
3. Let Bazarr extract the embedded subs

Software (please complete the following information):

• Bazarr: 1.1.0
• Radarr version: 4.1.0.6175
• Sonarr version: 3.0.8.1507
• OS: Linux-3.10.0-1160.11.1.el7.x86_64-x86_64-with

[vitiko98]

Embedded Provider won't touch any subtitles. This issue is related to post-processing logic.

[anonyme123]

Ok so the embedded provider just extracts the subs as-is, and then the post-processing logic runs:

• encode to UTF-8
• remove hearing impaired
• OCR fixes
• common fixes
• fix uppercase

so one of those is removing the 'SPAN' word?

[vitiko98]

Yes, I assume the hearing impaired processor is making the change.

[anonyme123]

Will you be able to fix it?

[vitiko98]

I'll try.

Please send the untouched subtitle file if you can.

[anonyme123]

Here you go: https://demo.lufi.io/r/AS3jQ9Mc6z#V2fXl71NM5mEZTQYK/AJTRsBYIFIVHC/VBy5pcJaIrQ=

[vitiko98]

Should be fixed in upcoming beta. Thanks!