Talking Book: Bringing Knowledge to Remote Rural Villages
Five years ago, we set out to design technology to provide on-demand access to oral knowledge for the poorest people in the world, who cannot read. We did this because we discovered how powerful knowledge can be to farmers living on $1 per day who want to learn how to produce enough food to feed their families and how to protect their children from deadly diseases.
Local experts, such as agriculture advisors and nurses, understand the challenges and know practical solutions for the poorest people living in their region; they also speak the local language. The challenge is making their knowledge available when it’s needed in the remote rural villages that are home to hundreds of millions of the poorest people on earth, who cannot read and often lack access to electricity. To address this challenge, we developed the Talking Book—an audio computer designed for the learning needs of oral people. A recently published evaluation of the Talking Book program showed that farmers using the device to access agriculture knowledge had a 48 percent improvement in crop yield, compared to a 5 percent decrease by their peers without access.
To reach our target unit cost of less than $15 and to improve the robustness of the Talking Book, we designed an interface that uses spoken audio instructions for output, rather than a display. For input, the Talking Book includes a set of buttons for users to press in response to the audio prompts, somewhat similar to the interactive voice response (IVR) systems that most of us have used through telephones. However, during our earliest user testing, we learned that too many levels of hierarchy can be difficult to navigate if you’ve never had any formal education. Our current design prioritizes simplicity: the user presses one button to rotate through categories (e.g. heath, agriculture, and local stories) and another button to rotate through all the messages within that category.
The messages that we produced in 2008 were typically just brief lectures by a local expert, but we soon found we were better able to engage users with an interview format. We started each project by assessing the needs, resources, and current practices of people in the villages we served. We then used that information to guide the interview questions with a public health official or farming expert. This format works well with agriculture advice, but influencing some health behaviors can require a song, a story, or even a soap opera. To learn what messages are working and not working so well, we track usage data and collect user feedback recordings. We know how popular each message is and what people like or don’t like about the messages. Since the Talking Book allows for multiple-choice quizzes and other interactive content, we can use these formats to reinforce key messages.
We have periodically considered whether mobile phones can adequately serve our objective to make on-demand knowledge accessible to the world’s poorest. The basic mobile phones that some of the poorest are able to access only offer the possibility of text messages and per-minute charges for voice calls through an interface that is far from ideal for the learning needs of oral people. For instance, listening and learning in groups is a common scenario as is listening while working on the farm; without a strong loudspeaker, this scenario isn’t addressed.
This project has been possible through a combination of donor sponsorships of villages and government partnerships. Groups of coworkers and friends have joined together to raise $5000 to sponsor the capital investment needed to bring the Talking Book program to a village of 1000 people. Partnerships with ministries of health and agriculture allow us to strengthen and extend their existing programs while continuing to keep the village programs operationally funded. Through this approach, we have impacted 20,000 people; and last year, we committed to President Clinton and the Clinton Global Initiative that we will reach 100,000 people within the next three years.