Speech technology is very useful and is likely to become increasingly widely deployed. As such, this area represents an important opportunity for Java programmers. Java speech synthesis technology is very mature, and the available tools are easy to use. A design area of importance is the way in which speech is incorporated into software. If machines are sympathetically equipped with this technology, the end user should be well pleased, especially given the way in which extra information can be added to the experience.
The FreeTTS package also indicates that speech synthesis has moved beyond the tinny voices of old to embrace a very functional feature set. You can easily record and incorporate your own voice into such software solutions.
There is a significant political element to the widespread deployment of this technology, however; well-designed speech-enabled solutions could eliminate a great many jobs. Take this factor into account.