This blog is now available as a podcast. I’m experimenting with the Odiogo service to automatically create an audio version of the blog text. The speech quality is surprisingly good, much better than the Windows speech synthesizer.
You can listen from the web page or subscribe to the podcast via iTunes etc. Let me know what you think.
Update: The speech quality is good on ordinary prose. It even pronounces foreign names correctly. Unsurprisingly, it mangles equations and special symbols. It generally has good inflection. It handles punctuation well, but it does not pause for <br> elements and so lists are unnaturally concatenated together.
I don’t know whether the podcast improves accessibility. I’d be interested to hear whether listening to the podcast is preferable to using a screen reader.