Extract MKV subtitles

extract_subtitles.ps1

# This script searches for all MKV files in a folder and its subfolders,
# then extracts the first English subtitle track found.

# ---------------------------------------------------------------------
# CONFIGURATION (EDIT THE PATHS BELOW)
# ---------------------------------------------------------------------

# Path to the folder containing your MKV files
$mkv_folder = "\\truenas.local\Plex\Series\Example"
# Path to the folder where extracted subtitles will be saved
$output_folder = "\\truenas.local\Plex\Series\Example\Subtitles"
# Full path to mkvextract.exe
$mkvextract_path = "C:\Program Files\MKVToolNix\mkvextract.exe"
# Full path to mkvinfo.exe
$mkvinfo_path = "C:\Program Files\MKVToolNix\mkvinfo.exe"

# ---------------------------------------------------------------------
# SCRIPT START
# ---------------------------------------------------------------------

Write-Host "Starting English subtitle extraction..."

Get-ChildItem -Path $mkv_folder -Recurse -Filter "*.mkv" | ForEach-Object {
    $mkv_file = $_.FullName
    $base_name = $_.BaseName

    Write-Host ""
    Write-Host "Processing file: '$mkv_file'"
    
    # Execute mkvinfo to get subtitle tracks and their languages
    $mkv_info = & $mkvinfo_path "$mkv_file"
    
    $track_id = ""
    $language = ""
    $codec_id = ""
    $is_subtitle_track = $false
    $found_eng_subtitle = $false

    # Loop through mkvinfo output to find the English subtitle track
    for ($i = 0; $i -lt $mkv_info.Count; $i++) {
        $line = $mkv_info[$i].Trim()

        # Find the track ID at the start of a new block
        if ($line -like "*Track ID for mkvmerge & mkvextract*") {
            $track_id_match = $line | Select-String -Pattern "track ID for mkvmerge & mkvextract: (\d+)"
            if ($track_id_match) {
                $track_id = $track_id_match.Matches.Groups[1].Value
                $language = ""
                $codec_id = ""
                $is_subtitle_track = $false
            }
        }
        
        # Look for "subtitles" to identify the track type
        if ($line -like "*Track type: subtitles*") {
            $is_subtitle_track = $true
        }

        # Look for the language
        if ($line -like "*Language: eng*") {
            $language = "eng"
        }
        
        # Look for the Codec ID
        if ($line -like "*Codec ID:*") {
            $codec_id_match = $line | Select-String -Pattern "Codec ID: (.+)"
            if ($codec_id_match) {
                $codec_id = $codec_id_match.Matches.Groups[1].Value.Trim()
            }
        }

        # If all information for an English subtitle track has been found
        if ($is_subtitle_track -and $language -eq "eng" -and $codec_id -ne "") {
            
            # Map the Codec ID to the file extension
            $extension = ""
            switch -Wildcard ($codec_id) {
                'S_TEXT/UTF8' { $extension = 'srt' }
                'S_TEXT/SSA' { $extension = 'ass' }
                'S_ASS' { $extension = 'ass' }
                'S_TEXT/ASS' { $extension = 'ass' }
                'S_VOBSUB' { $extension = 'sub' }
                default { 
                    $extension = 'dat'
                    Write-Host "  - WARNING: Unknown subtitle format ($codec_id). Using '.dat' extension." 
                }
            }
            
            $output_file = Join-Path -Path $output_folder -ChildPath "$($base_name).eng.$($extension)"
            Write-Host "  - English subtitle found (ID: '$track_id', Format: '$codec_id')"
            Write-Host "    - Extracting to: '$output_file'"
            
            # Execute mkvextract to extract the subtitle
            & $mkvextract_path "$mkv_file" tracks "$($track_id):$output_file"
            
            $found_eng_subtitle = $true
            break # Exit the inner loop after finding and extracting the first subtitle
        }
    }
    
    if (-not $found_eng_subtitle) {
        Write-Host "  - No English subtitle track found in this file."
    }
}

Write-Host ""
Write-Host "Subtitle extraction complete."
Read-Host -Prompt "Press Enter to exit"