Please let us know from which country you are having the bug (YT and Netflix have different content and subtitles depending on the country the user is located)
United States
If relevant, please let us know what are the configuration of languages you try to have
Italian > English
Looking at the network requests in Chrome, I see 400 status errors requests to:
Where do the transcriptions come from? I imagine they are not usually provided by the podcast producers themselves.
Hmm it’s a good question. I have to imagine they are auto-generating them, as it’s doing so with timestamps…the auto-scrolling seems to work, which you’d need the timestamps for.
FYI until this is fixed, my workaround is to export a transcript of my podcast from Youtube’s closed captioning (which are also auto-generated in my case), I used this tool (saw in another post) to do so:
I don’t love the export’s formatting, so I pasted the text into ChatGPT and had it fix capitalization, punctuation, and paragraph breaks.
Finally, took this and imported it into the “Texts” feature of Language Reactor. In this case, the translation feature does work for me.
You don’t get auto-scrolling, but it’s good enough for me, for now.