There are plenty of applications that can read off printed text. Windows has had that capability available in its setup for several years that I can recall now. And for the visual impaired that’s extremely helpful. There’s even a voice recognition module that lets you control things by speaking to it, again also useful for those with limited mobility for example.

Unfortunately the most difficult problem of all, creating text based on an audio file, still seems to be a ways off. I have found various sites that are researching it, and developing various applications, but nothing available for general use, really. I would like something like this in order to get transcripts from podcasts and such.

I was intrigued by this blog entry I spotted, discussing the recording of classroom lectures for later review. In particular, it sounds like it does at least some partly automated captioning, but it wasn’t clear from the description. That would be extraordinarily useful, an application that could do its own captioning as it recorded. But I can’t believe it does a complete job on its own? There must be another pass under a human to clean it up?

Would love to know more, since that implies the possible availability of general speech to text is coming up soon. And then I can wave a big fat finger at everyone who doesn’t provide transcripts, cause I’d no longer be dependent on someone else’s goodwill…or someone else’s willingness to follow the law…