{"id":27737,"date":"2024-05-23T14:27:08","date_gmt":"2024-05-23T12:27:08","guid":{"rendered":"https:\/\/www.graviton.at\/letterswaplibrary\/how-to-scrape-subtitles\/"},"modified":"2024-05-23T14:27:08","modified_gmt":"2024-05-23T12:27:08","slug":"how-to-scrape-subtitles","status":"publish","type":"post","link":"https:\/\/www.graviton.at\/letterswaplibrary\/how-to-scrape-subtitles\/","title":{"rendered":"How To Scrape Subtitles?"},"content":{"rendered":"

<\/p>\n

\n

There is very little Irish language text, audio and english translation. One of the best sources is this soap opera<\/p>\n

https:\/\/www.tg4.ie\/en\/player\/play\/?pid=6352950048112&title=Ros%20na%20R%C3%BAn&series=Ros%20na%20R%C3%BAn&pcode=669535&genre=Drama<\/a><\/p>\n

It is fairly easy to find the url of the subtitles when on that webpage manually<\/p>\n

getting the vtt file<\/a><\/p>\n

But the vtt URL uses UUIDs that seem pretty random<\/p>\n

https:\/\/redirector.playback.eu-west-1.prod.deploys.brightcove.com\/v1\/1555966122001\/7b5d6364-47e2-4016-ae63-93301a7f4e38\/ff7182e5-8f90-4af9-8d35-41a3bae7fa1e\/441366d1-6c40-4106-9c0f-ecfdc21476b0.vtt<\/a><\/p>\n

https:\/\/redirector.playback.eu-west-1.prod.deploys.brightcove.com\/v1\/1555966122001\/83680fe1-8055-4494-96ff-bc2786f937cc\/652c30ad-ff11-45d4-9e0c-46db42f5a34c\/0ab149e4-25b0-4c73-8c9a-8130d647de91.vtt<\/a><\/p>\n

There are subtitle archive sites but this soap opera is not there. So how would you extract a few hundred sets of VTT files (I want to build NLP datasets , ngrams etc, not make money or anything).<\/p>\n

I can imagine answers of<\/p>\n

With this site you can hire someone and if you show them the steps they can extract them for you cheap<\/p>\n

With this mouse emulator you can do it by XYZ<\/p>\n

There is away around the UUIDs being random by XYZ<\/p>\n

But I do not know how any of these would actually work.<\/p>\n<\/div>\n

submitted by \/u\/cavedave <\/a>
[link]<\/a><\/span> [comments]<\/a><\/span><\/p>