The real issue with current computer interaction is that touchscreens and keyboards still chain us to the device by our hands to the point where we are their slaves; Emancipated from those bonds, we will be able to reverse the roles back to where the computers are OUR servants.
It is clear that in humans, both sending and receiving voice communications can also be carried on to some extent alongside receiving computer video output, allowing at least some degree of multitasking. For example, I can listen to my "better half" while I blog! Voice to computer input is also faster for the non-expert than typing , and again we can do it in parallel with other activities. We can use voice simply for a limited range of understandable commands, or go further, into the ad-hoc dictation of typed information. So in the appropriate environment, voice input and output is an attractive option.
Products like * Dragon NaturallySpeaking and Apple's * Siri are the latest in a progression of the speech recognition technology, finding a natural home on mobile devices where a keyboard is at best a necessary impediment.
( * possibly the one and the same, as Apple wont confirm or deny
While speech recognition systems have fallen into disfavour in the contact centre environment (IVR systems) due to an unacceptably high error rate (on average 30% but sometimes much higher) , their use in a siutation where the user can train the system in the recognition of their own voice yields recognition rates of 90%. Industry reviews of the software have been positive .
No comments:
Post a Comment