Hearing Impared Subtitles

[Protektor-Desura]

Subtitles with hearing impaired support are not being imported properly. Actually it appears this isn't how Plex, Emby, Kodi, Jellyfin or Bazarr are doing things.

https://support.plex.tv/articles/200471133-adding-local-subtitles-to-your-media/ https://forums.plex.tv/t/hi-cc-sdh-subtitles-naming-convention/627100 https://emby.media/community/index.php?/topic/90668-external-subtitles-naming-convention-for-hearing-impairedccsdh/ jellyfin/jellyfin#4291

It is supported like this: Movie.en.hi.srt Movie.en.cc.srt Movie.en.sdh.srt

[mdhiggins]

Aside from missing the .cc and .hi is there some functionality issue with the import or are you just asking to support these extensions as well?

While the forum posts requests .hi this will break the language detection for Hindi and it appears Plex doesn't document .hi in their official support article but it would be easy to at least include .cc

[Protektor-Desura]

I am asking so that these files downloaded by Bazarr and others can be automatically imported properly in to the video file when I run SMA against them to clean everything up, tag them and standardize them.

[Protektor-Desura]

Also looks like "forced" can be one of the modifiers as well.

[mdhiggins]

All of the standard FFMPEG dispositions are recognized and tagged which includes

    DISPOSITIONS = [
        'default',
        'dub',
        'original',
        'comment',
        'lyrics',
        'karaoke',
        'forced',
        'hearing_impaired',
        'visual_impaired',
        'captions',

It sounds like you just want additional flexibility with sdh/cc/hi

It seems like emby/JF are only requests from the bazarr developer (not yet implemented) and plex didn't actually end up supporting 'hi' due to language conflict with Hindi but I'm looking into a solution

I was just trying to clarify if you're seeing a technical problem with the current import system which supports disposition analysis or if you just want this additional nomenclature recognized

[mdhiggins]

Well either way, looks like the old system was working on my end but I rewrote some of it to make it more robust, less false positives, language validation prioritized to avoid the Hindi issue, and added a system of alternate naming schemes to include the following additional matches

    ALTERNATES = {
        'comment': ['commentary'],
        'forced': ['force'],
        'hearing_impaired': ['sdh', 'hi', 'cc']
    }

Also fixed an issue where if the title of the movie file (which the subtitle had to include as the beginning portion of its filename happened to coincidentally contain one of the above dispositions it would falsely always match for that disposition, should not only analyze areas of the filename that are part of the suffix specific to the subtitle

Please try and report back

1e78208258982138fe10646066cf1079014c0ef4 a40d777d73fbe6eda57e4793bafcefba5580c9a0

[mdhiggins]

Closing this out due to inactivity, feel free to comment still