Downloaded audio continued its multi-year growth track. The category hit record growth in both units (27.0%) and revenue (26.8%) over 2013. While the category has been growing steadily and significantly, physical audio slightly declined. (source: dbw 2015)
The popular wisdom regarding eBooks, and what some see as their inevitable replacement of physical books for the majority of readers, is that eBooks are far more convenient than eBooks and therefore a superior format for the content, whether for entertainment reading or information and learning. The logic of that claim is, of course, that if a more convenient format arrives it will in turn replace eBooks. eBooks would then fit into the same niche market that current digital visionaries see the physical book retreating too. It is too soon to say, and there is not nearly enough data available to make a prediction, however the rising popularity of audio books and the recent lack of growth in eBook sales suggests this may be happening.
An audio book provides the same content as an eBook. It requires no unusual or cumbersome technology to use — mp3 players are now ubiquitous. It can be consumed while driving a car, skiing down a slope, walking the dog, or lying peacefully in bed. When I was a child my parents played records of classic literature (I remember Alice, the Three Musketeers and Robin Hood especially).
The technology is not yet there to make audio editions of books as easy to produce as eBook ones. While some eReaders may have an audio mode, the voices that read the text are far from pleasant and have no sense of the meaning of the words. That is changing, however, as semantic modeling develops. Researchers like Saif Mohammad at the Institute for Information Technology in Ottawa, Canada are already mining literary texts to map out the emotional content. Below is an example of an analysis of Shakespeare’s Hamlet in which Fear and Sadness are far more common than Trust or Joy. (This may come as no surprise to anyone who knows the play).
Once the emotion of content can be identified, the algorithms for adapting the reading voice to match those emotions become possible. Add on to that semantic understanding of the context of that emotion and you start to have a powerful, or at least enjoyable, reading performance. I suspect shortly after that the algorithms will be adjusted to provide a variety of personalities, responding to the emotions in a variety of ways ranging from sympathy to cynicism.
And that leaves the eBook having to contend with the physical book for the reading market that does not want to be read to. The big book chains were meant to drive the independent booksellers out of business, but they then all but disappeared themselves when Amazon arrived while the independents remained (providing a service Amazon can’t). Perhaps the eReader and eBook will face a similar future. Let’s face it: its hard to replace a piece of technology that has been working well for centuries.