Just as a note, recently I’ve been talking to a lot of people about nonstandard interfaces. For the next bit I’m willing to test out one of the current “nonstandard interfaces” available to us right now by dictating my blog posts using the native Windows 7 voice interface. I’ll be using the keyboard as little as possible during this test to determine just how far we’ve come with voice recognition. I doubt I’ll ever abandon the keyboard completely as that was my first interface to computers. But with this experiment I’ll be testing voice recognition in a variety of environments and tasks to determine if I can extend my toolbox for interacting with my systems. With two blog posts dictated already using voice recognition, so for I’m very pleased with how it’s going. The fact that seven is almost always spelled out and I have to correct it to a number is my biggest gripe speaks well for where we are.
I imagine the true test will be dictating in my car. That’s a notoriously noisy environment as my built in voice recognition system for my hands three phone can attest. But if it can work there even somewhat reasonably, then I’ll be able to take some of my input time (podcast listening) and turn it into input/output time (the reason I like to listen to those podcasts is that I get some great ideas, and having a better recording medium is always very helpful.)
