I am using gTTS in Python and I am having the problem of it stuttering and repeating, as in the attached.
I am calling it like:
tts = gTTS(text)
tts.save(path)
as is normal. I have tried generating a different tts using the same phrase as the glitchy ones and the new mp3s are correct, so this appears to be a fluke of some kind. However, it is common enough to make it effectively useless unless I can fix this bug. When looking at the raw data in the file I see a large number of 'U's and I'm not sure what this means but these seem to be in the working tts files as well. This also happens when I make an API request to the TikTok TTS API with a similar frequency and appears to increase in frequency with longer strings. I was wondering if there was a way to detect these strange errors in order to regenerate the needed tts segments or avoid them entirely. Thanks!
