These two phrases are surprisingly similar when broken down into phonemes, yet have very different meanings. A program using a Markov Model at the sentence level might be able to ascertain which of these two phrases the speaker was actually using through statistical analysis using the phrase that preceded it. When the computer checks its Markov Model it would find that the first phrases is much more likely to appear than the second, so it uses as a base for the rest of the sentence. Yet, if the phrases where these sentences:
James's school drives monkey feet
James is cool because he plays the guitar
It would switch over to the second sentence, because the first sentence quickly turned meaningless, while the second was both semantically and syntactically correct. This sort of "understanding" is imperative for a speech recognition engine, and especially one that has to deal with continuous speech.
…
