After a great start into Voice UI and conversation design, the feedback and the mixed bag of opinions began to flow my way. This typically means "Dig deeper" for me. So i spent the last couple of months exploring the foundational elements of Voice services
1. Computer Sciences
2. Artificial Intelligence
3. Linguistics.
I feel the deep dives really helped focus the problem domain for VUI by understanding the roadblocks to conversations. Contrary to the connotation of NLU- there is no real understanding of voice. We knew that already, but our mental model made it hard to work around the fact. If the NLU does not understand, what does it do? What am i using?
VUI uses the NLU as a improving question and answer system. For a good mental model, think of NLU as a system that helps answer questions. You as a VUI expert get to define the framework to define questions and answers.
To build a great conversation, you have to use a series of questions and answers to simulate a dialog.
That is it.
I am back in the saddle and expect several bits of experiences in the coming weeks.
1. Computer Sciences
2. Artificial Intelligence
3. Linguistics.
I feel the deep dives really helped focus the problem domain for VUI by understanding the roadblocks to conversations. Contrary to the connotation of NLU- there is no real understanding of voice. We knew that already, but our mental model made it hard to work around the fact. If the NLU does not understand, what does it do? What am i using?
VUI uses the NLU as a improving question and answer system. For a good mental model, think of NLU as a system that helps answer questions. You as a VUI expert get to define the framework to define questions and answers.
To build a great conversation, you have to use a series of questions and answers to simulate a dialog.
That is it.
I am back in the saddle and expect several bits of experiences in the coming weeks.
