Apple's Macintosh Operating System has always been committed to speech technologies. The first Mac spoke with a speech synthesizer when it was first introduced in 1984. The Mac was the first personal computer to ship with speech recognition built in. Since then Apple has continued to develop its own speech and language technologies. In this presentation I will systematically describe and demonstrate the speech technologies that are built into "Jaguar", the latest release of Mac OS X. These include speaker-independent speech recognition, text-to-speech synthesis, latent semantic analysis, and an array of general user interface features that use these technologies to do useful things for the general user. I will describe our current understanding of the role of speech in personal computing, motivate some of the design issues we have made, and illustrate some of the problems that are still to be solved.

Target audience: general users of personal computers, human interface designers, speech technologists.