Archæology

The assorted finds of Artefact Publishing

Text to speech engines

You would think — or rather, I think — that if you’re going to make a text-to-speech system, you should base it on the IPA. That is, after all, its entire point. And indeed, various efforts in providing specifications for how to go about the task have included ways of including IPA information. However, the actual systems people have built don’t seem to use them. That is silly.

However, this has been recognised and some people have hope:

Unfortunately, IPA is difficult to learn and use and there is not yet standardization on the use of subsets of IPA for particular languages and dialects. Nevertheless, speech technology is converging on IPA as the only available system to represent the sounds of a wide range of languages and dialects. There is some hope that this convergence will lead to developer tools and increased standardization that will make IPA more practical.

So, in a similar manner to how computer people eventually caught on to the fact that librarian’s have been doing “metadata” for a long time and might know something useful about it, perhaps they’ll soon come to see that there’s merit in doing what beginning linguistics students do as part of a single course.

Posted by jamie on June 2, 2007 14:48+12:00

Comments