I upload a spoken Cantonese audio file in MP3 and wait as it tells me, but it never successfully generates anything. I heard LingQ uses Whisper for the transcript generation, and so I went and put it through Whisper directly myself on Google Colab, and there I have it: a complete transcript in a few seconds. I asked a friend to confirm it’s accuracy, and it was an accurate transcription.
I appears LingQ can’t handle this. I am disappointed as this feature is one that brought me back here after a long break. But each time I try to get this working it fails, and I just waste time troubleshooting when I wanted to be studying!