Today’s twitter feeds stumbled me about this awesome article on “whatever happened to voice recognition”.

Having tried and used speech recognition myself (all started on my 400 mhz celeron … oh boy…) i really agree on the article’s essence. They just don’t get it right – and it’s more or less impossible to really get it right.

There’s also some current state-of-the-art software available. Tried it, failed. It’s not that it doesn’t work at all, but that you are faster typing than speaking, going to keyboard fixing, speaking again, hoping and checking, fixing….

Why on earth I am thinking now on the big great Android feature to fill *every* textfield with text-to-speech via google servers? I don’t know. Maybe it relates to this. (And yes, I tried it, and yes, it failed in a quite funny way. Maybe because I’m not a native speaker… maybe.)

I wonder how many people actually like this feature. I’ll add a comment here as soon as I see someone non-geeky using it. And don’t bash me now – there’s a difference between voice control and voice recognition. In short, the first may work

So again, it’s great for navigation, or other feature sets where you have a finite domain of answers. But it just fails on free speech.