Issue with unicode characters when encoding a subtitle file

[Hemer]

Hi,

I'm using yt-dlp to get videos + subtitles from arte.tv, then watch them on my TV using Universal Media Server.

It seems there is some issue with the unicode characters of the subtitles I get cause all (french) accented chars are not displayed correctly by Universal Media Server:

• é => e´
• ê => e^
• ç => c¸
• ö => o¨
• etc.

All ​punctuation marks and other special chars also look weird (like not as usual).
Note that subtitles are correctly rendered by VLC.

You can reproduce the issue by encoding the joined file to Windows-1252, ANSI, ISO-8852 or any other european compliant encoding (this is what I want actually: converting without the glitches to ensure a proper rendering through UMS).

When I change encoding using Subtitle Edit I get an alert "Subtitle contains Unicode characters. Saving using ANSI file encoding will lose these".

I could fix the issue by using some find & replace method, but maybe you guys got a better (and less boring) idea?

I've tried a lot of unicode / encoding converters, online and local (Notepad++, Sublime text, Subtitle Edit). Also tried different ways to recover the subtitles from the source.

Thanks.

[Hemer]

So, I found out that the original subtitles I get from arte.tv are using unicode precomposed characters.

This thread gives several solutions to fix the problem (including an online unicode encoder/decoder).