The pre-cursors to mobile phones were two-way radios, also called Walkie-Talkies, that transmitted voice signals. The first generation of mobile phone networks were similar in that they also only supported voice communications. Second generation networks, and a happy accident, gave us SMS, and third generation networks provide even more advanced mobile data services.
Most mobile phone applications now use these newer channels of communication — SMS and data. But even though we sometimes forget, voice is still a major part of mobile phone communications. And when it comes to performing social work, voice communication is actually the most important feature in many parts of the world. I’m going to profile some of the most interesting of these initiatives, but first it’s important to understand why voice is playing such an important role.
Voice has a big advantage over SMS and data transmissions because it channels spoken language directly. People of varying literacy levels are able to use voice technology with keypad and voice navigation, and this means applications can be run in local languages. Users can issue commands and requests and thus communicate more effectively.
The downside of using voice comes on the receiving end. Voice data is much harder to process than text or other data. It requires considerable technical effort (or a lot of people power) to parse and separate voice data. Even then, the accuracy isn’t perfect. Searching through voice data also remains a near-impossible feat. On top of that, airtime costs tend to run higher with voice than for text message.
Yet, even with these challenges, there are several notable projects that leverage the voice capabilities of mobile phones to deliver important information in interesting ways.
Question Answering Services
Two of the more notable projects provide a very simple service: they answer people’s questions.
Question Box provides a service in India and Uganda. In India, phone boxes are installed in slums and villages that connect users to operators that will answer questions. In Uganda, users can call in from any mobile phone and ask their questions. The operators have access to a repository of previously asked questions (and their answers), and they can also occasionally consult the Internet. A special search engine and database were also built specifically for the project.
Another initiative, Avaaj Otalo, provides an audio community forum for farmers in rural Gujarat, India. Working with an organization that produced a popular radio program, Otalo provides a call-in number where farmers can exchange questions and answers. Users are also able to listen to archives of the radio program.
These projects differ in that Question Box avoids having to process users’ questions by adding a human listener in the loop; Avaaj Otalo avoids processing by organizing their collection of audio prompts with into a menu. Both programs, however, have yet to deal with the problem of cost because they subsidize the service for users. Otalo operates with a toll-free number and Question Box provides the phones to call from in India. In Uganda, Grameen Community Knowledge Workers provides the mobile phones.
Wikipedia and News on the Phone
MobilED, which operates in South African schools, developed a program that delivered Wikipedia content via mobile phones. Users texted in a query and were called back by a speech synthesizer that read the text of the relevant Wikipedia entry. Users could also upload voice-based edits to articles, or create audio entries if nothing existed on a given topic. The queries were easier to handle because they were text-based, but the information delivery utilized voice in order to deliver the entry and ensure comprehension. Unfortunately, the service was expensive to maintain and MobilED eventually abandoned the project in favor of purely data-based services.
FreedomFone, a Knight-funded project based in Zimbabwe, is working on providing news using an audio channel. In an environment where the press is highly censored and access to news is scant, FreedomFone plans to implement a solution so users can either call or text and receive the latest news information. The cost structure has yet to be determined.
Recreating the Web, or Wikis Over Audio
Perhaps the most ambitious project is IBM’s Spoken Web (also known as the World Wide Telecom Web) Project. The idea is to create an all-audio version of the web. The project aims to create voice-sites that are linked to specific phone numbers, much in the way a website has a URL. The project has already built a system called VoiceGen that creates VoiceXML content from an audio input. The group deployed part of the technology in the form of VoiKiosks, a service that allowed users to listen to information from different NGOs and upload professional advertisements. It proved to be a popular test service.
The Computer Science and Artificial Intelligence Laboratory at the Massachusetts Institute of Technology is also developing audio wikis, or “local repositories of audio information,” that can be edited and created using audio. It essentially recreates wikis on a mobile-accessible audio platform. The system is not fully developed, but is expected to be deployed in India in conjunction with Microsoft Research India.
The barriers to voice services
Voice communication offers several benefits, especially when it comes to low literacy consumers. The question, then, is why voice-based technologies are not more of a player in the world today? The automation of these types of service is difficult, but Question Box, Avaaj Otalo, and IBM have shown that challenges can be overcome.
Cost seems to be the biggest issue when it comes to deploying and maintaining these services. MobilED ended its voice-based program because of cost. It’s important to note, however, that airtime is particularly expensive in South Africa. The International Telecommunication Union reported in 2008 that costs for one minute of on-network airtime during peak hours was the equivalent of U.S. $0.59 when calculated using purchasing power parity. Airtime is much more affordable in India. The same minute in India costs only U.S. $0.07. Hopefully, this will encourage other voice-based mobile services in India, and we can eventually see the cost issue resolved once and for all.