Kindle Text-to-Speech: “Robotic NPR”

Photography imported on the site Flickr.com by...
NPR’s next hire? (Image via Wikipedia)

I’ve made a discovery about Amazon’s Kindle e-reader: It’s a pretty good “news radio.” That is, its text-to-speech function does a surprisingly decent job of reading news content aloud.

I currently subscribe to the Wall St. Journal on my Kindle, and I’ve gotten in the habit of letting it read me some interesting articles as I go through my morning routine. I like it. The automated text-to-speech reader is a bit flat for fiction, narrative, and essays that require significant emotional or rhetorical inflection — but it’s great for news. I’ve starting considering it my “robotic NPR.”

(Ducking the reflexive outcry from all my friends at NPR…)

Of course, my point isn’t only about the Kindle. It’s about how any text-to-speech service or tool can interact with text-based news and information content — and why creators of text-based news content should start to take that into consideration. Because you never know exactly how people will experience your content…

Like e-reader display technology, text-to-speech technology has improved significantly in the last few years. It’s still far from perfect, but of all the versions I’ve heard the Kindle’s is one of the clearest, and others are catching up. This is good for people who have a preference for audio news, because now we can experience news produced for text in a format that works with our preferences.

Don’t get me wrong — I love news specifically produced for audio (either radio broadcast or audio/video news podcast). I listen to a lot of it. (Oh, if you haven’t tried the Public Radio Tuner iPhone application, get it, it’s killer.)

Still… It’s pretty cool to be able to have stories from WSJ.com read aloud to me while I cook my veggie pesto omelet. Or articles from the newly online-only Seattle Post-Intelligencer, which I can quickly “Kindlfy” via the free Instapaper service, which I wrote about earlier.

As text-to-speech technology continues to improve and proliferate, I’d suggest that text news publishers consider how well their online and Kindle content “reads,” in the audible sense. One thing I don’t like about listening to WSJ stories via Kindle is that it reads aloud all the navigational context at the top of the story: word count, etc. This is just a minor and fast irritation, but it bugs me. There’s got to be a way to get around that.

So, as I recommended when I first wrote about the Kindle 2, when your newsroom gets its Kindle (or when you get to fondle someone else’s for a bit), try listening to some news stories (preferably your own, but anyone’s news is a good start). You can subscribe to many newspapers and magazines via the Kindle store for a free two-week trial, or buy an individual article or two. Play with the settings for speed, gender of voice, etc. Realize that you’re listening to a stepping stone technology that presages a potentially important channel for your news in the future. And just keep it in mind.

(Note: This is a slightly re-edited version of an article I originally published in Poynter’s E-Media Tidbits.)

Reblog this post [with Zemanta]

3 thoughts on Kindle Text-to-Speech: “Robotic NPR”

  1. Pingback: An Unnecessary Barrier | Universally Designed

  2. Pingback:   What’s Going on with WSJ Pricing? — contentious.com

Leave a Reply

Your email address will not be published. Required fields are marked *