Do voice applications requiring decisioning?

I saw this interesting article on Computerworld - Can You Understand Me Now? It seems to me that the growth in voice recognition is only useful if you can decide what to do about it. Using voice recognition simply to navigate menus seems pointless. Surely the purpose in improving voice recognition is to enable people to get decisions made and actions taken using only their voice? It seems to me that this is a separate problem from the voice recognition problem.

So, if I can accurately parse the words, the meaning and even, according to the article, the emotion of a person you will still need a rules management system to make sure that you respond the right way to what the person asks.

If speaker is an_annoyed_speaker and speaker's requests include travel_request then next_action is refer_to_csr_supervisor

The article makes a great point that constant change and the ability to modify how the system responds is an issue with voice recognition. This kind of agility is a key benefit of business rules.

Just a thought...