We’re running an audio experiment

Starting on September 18, we’ll be running an experiment. Over the next four weeks, every story we publish will be accompanied by an audio recording of the text. That recording will be made with Amazon Polly. It’s a service that converts text to speech. What you will hear is a computer voice (we’ve chosen one named Ruth) that should sound very much like a normal human voice.

Why are we doing this? Well, we have a hypothesis. This is a statement that we can test to see if it’s true. Our hypothesis is that having audio versions of stories will improve the experience for you, the reader.

We know that some people can benefit from audio like this. Those groups include people who have vision impairments, those with dyslexia and individuals for whom English is not their native language. Also, there are plenty of people who just prefer to listen to words rather than read them, or like to listen and read simultaneously.

Our test will place an audio recording at the top of every story we publish for the next four weeks. During that time, we’ll collect data so that we can analyze that test when it’s done. Some of those data will be things like how many times people listened to the recordings. But a big piece of data will come directly from you. Sometime in the next four weeks, please take our survey. (It will be linked after every recording.) Tell us what you think. The good, the bad, we want to hear it all.

After we have all of these data, we’ll run an analysis. If the data support our hypothesis, then maybe we’ll incorporate the findings of this experiment into how we run this publication. All thanks to you.