A little script to help with extracting subtitles from .mkv files.

subextract.sh

#!/bin/sh

if ! command -v mkvmerge mkvextract >/dev/null 2>&1; then
    echo >&2 "Tools mkvmerge & mkvextract are required. Install package mkvtoolnix."
    exit 1
fi

filename=$1

if [ ! -f $filename ]; then
    echo "File not found."
    exit 1
fi

filename_no_ext=$(echo $1 | sed -r 's/.mkv//')
sed_exp='s/.*([0-9]+):.*language:([a-z]+).*/\1:'$filename_no_ext'.\2.\1.srt/'

mkvmerge -I $filename | grep subtitles | sed -r $sed_exp | xargs mkvextract tracks $filename

Fails when there are more than 10 tracks, fails if the subtitles are vobsub (because mkvextract will then ditch anything beyond the first dot, and make two files named .idx and .sub respectively -- probably a bug in mkvextract, it should split off at the last dot, not the first).

Here's a sed-line that fixes the first and works around the second.
sed_exp='s/Track\sID\s([0-9]{1,3}):.language:([a-z]+)./\1:'$filename_no_ext'-\1-\2/'

And in order to make this work with filenames containing the path, you need to do this:

filename_no_ext=$(basename $1 .mkv)