After reading this article, I'm confused about Apple's strategy with respect to Speech. They were one of the companies that drafted the Java Speech API (JSAPI) in 1998. Of the original drafters of the spec, they are the only ones that have yet to provide an implementation.
Given that they are the only operating system that ships with Speech Recognition and Synthesis built directly into it, you would think that they would have connected the dots between their platform specific code and the JSAPI framework. What's the hold up?