PowerShell script to download and extract plain text from autogenerated subtitles of a YouTube playlist

DownloadYouTubePlaylistSubtitles.ps1

param (
    [string]$playlistID
)

if (-not $playlistID) {
    Write-Host "Please provide a playlist ID as a command-line argument."
    exit 1
}

# Define the playlist URL and output file
$playlistURL = "https://www.youtube.com/playlist?list=$playlistID"
$outputFile = "all_subtitles.txt"

try {
    # Fetch the list of video URLs and titles in the playlist
    $playlistInfo = yt-dlp --flat-playlist --print "%(id)s %(title)s" $playlistURL
    Write-Host "Fetched playlist information."

    if (-not $playlistInfo) {
        throw "Failed to fetch playlist information."
    }

    # Initialize or clear the output file
    "" > $outputFile

    # Split the playlist information into IDs and Titles
    $playlistLines = $playlistInfo -split "`n"
    Write-Host "Split playlist information into lines."

    if ($playlistLines.Length -eq 0) {
        throw "No videos found in the playlist."
    }

    # Loop through each video, download the subtitles, and extract the plain text
    for ($i = 0; $i -lt $playlistLines.Length; $i++) {
        $line = $playlistLines[$i]
        # Write-Host "Processing line ${i}: ${line}"
        if ($line -match "^(?<id>[\w-]+) (?<title>.+)$") {
            $videoID = $matches['id']
            $videoTitle = $matches['title'] -replace '[^\w\s-]', '' -replace '\s+', '_'
            $subFile = "$($i+1)_$videoTitle"
            # Write-Host "Video ID: ${videoID}"
            # Write-Host "Video Title: ${videoTitle}"
            # Write-Host "Subtitle file: ${subFile}"
            
            yt-dlp --write-auto-subs --sub-lang en --sub-format ttml --convert-subs lrc --skip-download -o "$subFile" "https://www.youtube.com/watch?v=${videoID}"
            
            $fileName = "$subFile.en.lrc"
            if (Test-Path $fileName) {
                Add-Content -Path $outputFile -Value $videoTitle
                Write-Host "Subtitle file ${fileName} downloaded successfully."
                $subContent = Get-Content $fileName
                foreach ($line in $subContent) {
                    if ($line -match '<font color="#ffffff">(.*?)</font>') {
                        $plainText = $matches[1]
                        # Write-Host "Extracted text: ${plainText}"
                        Add-Content -Path $outputFile -Value $plainText
                    }
                    else {
                        Write-Host "No match found in line: ${line}"
                    }
                }
                Remove-Item $fileName
            }
            else {
                throw "Subtitle file ${fileName} not found."
            }
        }
        else {
            throw "Line did not match the expected pattern: ${line}"
        }
    }

    Write-Host "All subtitles have been downloaded and merged into $outputFile."

}
catch {
    Write-Host "An error occurred: $_"
    exit 1
}