I just want to chime in here and say that I have been having fun experimenting with these features.
When selecting a specific voice (male voice Thomas FR) for text-to-speech on my Mac through Readlang works to play back the entire text with the Read Aloud feature, but it still plays a female voice for selected words or phrases during look ups.
Could you check whether this is a ChatGPT female voice or a different macOS voice providing the audio? The top voice in the list is a female voice so maybe it is defaulting to that for selections? Something is weird here…
Also, would it be possible to add an option to have the Read Aloud stop after each sentence? I like to hear the pronunciation of the sentence before I do my look ups for new text because French has so many silent letters and I am learning to identify patterns.
If it would be possible to pause native audio from an uploaded mp3 after each sentence that would be great too, but I assume that would require word-level timestamps or something so might be complicated to implement. For now, I would be very happy to have it as an option for the Text-to-Speech voices.
Thanks again.