Audio switches unexpectedly after text-to-speech generation

I’d like to report an issue I encountered.
While text-to-speech audio is being generated for one lesson, if I open another lesson,
the audio in the lesson I’m viewing is replaced by the newly generated audio once the generation finishes.
Reloading the page restores the correct audio.
It’s not causing major issues for me, so I don’t mind if it’s not fixed right away, but I wanted to bring it to attention.

Thanks for reporting! We will investigate the problem.