Hardly, of course, but that's what the story says. Readers who know German might still enjoy this link — A well-known Austrian actor, Hanno Pöschl (pictured on the left, image from here), has given his voice to science, or at least donated samples for use on "the world's first computer that speaks Viennese." If you scroll down a little on that link, you can hear the contrast between Pöschl actually speaking and Pöschl's synthesized voice.
Aside from being cute, it's a little reminder that speech recognition has to be able to deal with real-world linguistic variation in sounds, and it's even nice to have synthesized voices in varieties we prefer to speak/hear. Still, let's not think too much about how far this computer is from 'speaking', let alone producing actual Viennese.
Wednesday, June 25, 2008
Subscribe to:
Post Comments (Atom)
No comments:
Post a Comment