> Disfluency detection is the task of identifying self-repairs, repetitions, restarts and filler words in language. I would like to know more about how and why

> Disfluency detection is the task of identifying self-repairs, repetitions, restarts and filler words in language. I would like to know more about how and why this is used, and why it is so important that Google is building a model for it. As a layman, it sounds to me like it would be useful for writing suggestions or evaluating writing quality. Is that correct?