Despite what you see in movies, the reality is that computer voice recognition still sucks. There’s a great illustration of this in the following video…designer Michael Silber had Siri speak some text, and that was then sent to Google Voice, which transcribed it into text. That text was fed back into Siri..and the process repeated. It’s much like the ‘telephone’ game that kids play, and the results are similar…the message quickly degrades into gibberish.
Current state of computer voice recognition
Comments are closed.