Listening to your own paper
I’m sure many of you already know about the ‘Read Out Loud’ capability of Adobe Acrobat. I’ve been doing so much writing lately (a dissertation chapter) that I end up reading the same passages over and over again and can’t think of ways to fix problem areas or, worse, I miss glaring errors staring me right in the face.
So, i’ve started copying/pasting specific passages of text into a new document, converting it to a PDF, and then listening to it. Sometimes I think of new (once in a while, better) ways to say what I want to say. Other times, it’s not so fruitful. But….it’s another little tip that seems to help me produce better technical writing. The key is to only do a couple paragraphs at a time….the computer-guy voice (like from that Radiohead album) gets a little hard to listen to after a couple minutes.
The screenshot above simply shows where this functionality is in Acrobat.
If the voice was more realistic and easier to listen to, I would love to be able to download a paper as an mp3 and then listen to it on the train ride. Is anybody doing such a thing out there?
UPDATE: I found this ultra-easy web application that converts any text to mp3, which you can listen to right away or download to listen later. It’s called vozMe. I tested it, the voice is pretty similar to one I discuss above. Okay…but still not sounding very human…yet.
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Clastic Detritus 
There’s a feature on Macs that will read any selected text with a user-defined keyboard short-cut. Mix that with a sound recorder and you could turn any old paper into an MP3. This is also good, because you don’t have to convert anything to PDF, you just select and play from Word, or your Latex editor, or whatever you happen to be writing in.
But there’s problems I’ve found with technical documents. It can get very difficult to understand when Stephen Hawking-like voice is talking about “one hundred and eighty seven ree slash one hundred and eighty eight oss” when what he’s really trying to say is 187Re/188Os.
I’m afraid the voice will stop halfway through and say “hang on, you actually think this crap will get published? I mean, you are missing key references, are poorly organized, and man, really you aren’t concluding much of note!” Then it would tsk tsk me, and start reading aloud academic job rejection letters it found in my hard drive, laughing the whole time. Dangerous tool I say.
Yorrike….thanks, yeah it’s the unnatural voice that kind of gets to me….I’m hoping the journal Sedimentology hires Kelsey Grammar or David Attenbourough to do dramatic readings of the latest articles, and then release via podcast. I think that has a good chance of happening.
T-chronic…always important to stay aware of the perils of AI. That’s how the machines will take over, by slowly nagging and pointing out your shortcomings until you say “fine, do you whatever you want”