Blog posts to audio file on Mac OS

This is a wickedly cool solution to a common situation. Quite often as I’m skimming through Google Reader, I come across a tasty looking blog post that looks interesting but is too long to read through at the time. I could bookmark it in my delicious account, but I’m likely to forget to do this. I figure there are two solutions:

1. Hire someone to write me a short summary. I wouldn’t want to go this route, even if I could afford someone to do it, because I’d lose the full body of thought the author wanted to say. In Mac OS, this can actually be done automagically using the “Summarize” service (available from the application menu -> Services). It will create a reasonable summary of any selected text.
2. Hire someone to read the post to me. Better yet, hire someone to record a reading of it so I can listen to it while I’m doing something else.

The happy news is that there is a way to do the second alternative in the Mac OS, and if you aren’t afraid of using the terminal it is stupendously easy (it might be helpful to know some basic terminal commands to do this). I found this process courtesy of Ask Metafilter. Here’s how:

1. Copy the text of the blog post into a text editor. Save it as a text file and make sure you have the “.txt” ending. You could call it “blog.txt” for example.
2. Open the terminal utility. Change directory to the directory where you saved the text file.
3. Here’s what you need to enter to create the audio file:

   say -f speech.txt -o speech.aif

   The voice selected in your “Speech” system preferences will be used as the speaking voice in your audio file.
4. The file you create will be in “aiff” format. You can use iTunes to change to a different audio file format, such as mp3.

There are a few things that could be done to make this process even cooler (if that is possible). First, I could shell out some money for some better quality voices. The Alex voice that comes with OS X Leopard is pretty decent, but there are better ones available for purchase (such as those for sale from AssistiveWare. Personally, my top choice for a voice would be  StephenFry’s). Another good thing to figure out is to automate this process using Automator or AppleScript. If that could be done, it might be relatively simple to input an RSS feed then have the items put into an audio file or read out loud to me while I work. This things may already be possible but I’ve just been too lazy to ask The Google to find them for me.

Here’s an example. This is the converted version of an old post of mine “It’s a river, not a reservoir“.

[display_podcast]

Any suggestions for improving this process are definitely welcome.