Well, well. I figured with Christmas upon us it would be kind of a slow week for the blog, but I have figured wrong.
There’s been an outpouring of response to my post about the GirlTech Password Journal (GTPJ). So far, the post has seen more comments than any in Speech Tech Blog history. A number of disgruntled Speech Heads have been writing in to say that they’re dissatisfied with the biometric speech solution. I can’t say for sure, but it seems like many of you out there have been coming to us, having maybe received one of these bad boys (well…girls, really) for Christmas feeling like you got the short end of the stick and looking for answers.
The chief complaint out there seems to be that the GTPJ not only keeps offending parties from reading the journal’s most private thoughts contained inside, but also the authorized user.
One reader put it best when he or she wrote, “MY 6 YEAR OLD SIXTER HAVE ONE BUT IT RUBISH.”
I have to admit, I’m kinda glad that we’ve tapped into a raw nerve here. Not only has it gotten a lot more people looking at the blog (always good), but it seems to be highlighting, in its own roundabout way, a perennial problem in speech: if you don’t spend a lot of money, there is a lot of short end of the stick to go around.
Because of high development costs and the increased centralization and consolidation of major sectors of the speech industry through mergers and acquisitions, this speech stuff has stayed relatively expensive for a long time. For large enterprises that can absorb the big expenses, quality hasn’t been an issue. They can spend the capital to get speech solutions that work with a high degree of accuracy and precision. However, when you’re making toys and are looking to keep costs low and maximize profits, that’s another story altogether.
I remember my sister’s Furby, for instance. That thing recognized what we were saying only maybe 30 percent of the time—and that’s probably being pretty generous. To keep toymaking costs low, the hardware and software on many speech-enabled kid’s products has been less than thrilling. This is particularly egregious when manufacturers make all sorts of promises to kids that the products just don’t deliver. The stuff never seems work like it does in the commercials, and many a child is left bawling mercilessly as his parents tear their hairs from the roots in wild frustration.
This isn’t true all over speech. We’re starting to see some positive movement in price, especially with speech-enabled smart phone apps–a lot of which are really good and free to boot. But speech, like history, doesn’t move in a straight line. So watch out.
I’ve said this all before when I warned “Caveat emptor” at the end of my original post about the GTPJ, but maybe I ought to translate this time around so there can be no question: Let the buyer beware. Yes, buyer beware, dear Speech Heads.
