Feature Request: Visible Speech Pronunciation Feedback
When doing the speech segments, It can be quite frustrating trying to figure out which part of the sentence is not pronounced correctly. For some sentences, I've had to repeat it so many times, that I think it is actually broken or over capacitated. Other times, you have to use brute force, which is a time waster.
Not sure how the audio wave analysis is done and what type of algorithm is used; but I'm sure it would be possible to pinpoint the errors in pronunciation. This would be such an effective feature.
Maybe something like this: