Your boots hit the ground in the province of Aceh, Indonesia, the morning of December 29th, 2004. Several nights before, a magnitude 9.2 earthquake–one of the strongest ever recorded–struck off the coast of southern Asia, sending a series of massive tsunamis across Indonesia, Thailand, India, and other countries. You’ve been deployed in an emergency task to provide aid for the victims here in Aceh. But no amount of briefing could prepare you for what meets you on arrival.
The scene in front of you is one of devastation. The zone of destruction extends nearly three kilometres from the coast. The village here has been reduced to water-swept debris. The water that seeps out of the carnage and collects in pools is black with mud, gravel, and other pollution. A few dozen people mill about what presumably had been their homes a few hours earlier, searching for their belongings, their loved ones, anything they can find.
As you unload your supplies, a local villager approaches you, looking scared and confused, holding back tears. His speech is fast and low, in a language that you don’t understand. He gestures behind him urgently and tugs at your sleeve, and when it is clear you don’t understand, he bursts out into tears.
This is the scene that unfolded across not only Aceh, but over a dozen countries in the days following the horrendous natural disaster of 2004. As relief organizations from across the world mobilised to aid those in need–providing fresh water, food, medical aid, and burying the dead–they repeatedly experienced communication problems in the places they visited.
Resources were difficult to allocate, as relief workers could not easily determine what the victims needed most. As the debris cleared and more aid arrived, volunteer translators for less widely-spoken languages became more available, but even so, there are too few translators in the local language in these remote areas. For some, sadly, this was too late.
Now imagine instead, if when that local approaches you, tears in his eyes and a foreign language on his tongue, you can reach into your waist pocket and pull out your phone or a similar small device, press a button and begin to talk to each other through a translation system on the device, called a “Douglas”. Even though there is no internet connection and no human translator available, the Douglas records his voice and almost immediately begins translating his words for you out loud in English. His wife is trapped, he tells you, in the rubble of his family home. You speak into the device: “Show me where she is.” The device translates immediately into the local Acehnese language, and the man nods and leads you to her.
This future vision of humanitarian aid inspired the Humanitarian Babel Fish project, started by a small team under the project authority of The Warren Centre for Advanced Engineering within the University of Sydney. This international group has been working since 2014 to create a portable device for first-line emergency personnel in field areas where the task force may not have the means to communicate with the locals in the disaster zone.
The way the Douglas device works is simple. Each Douglas will be set up in advance with modular local language “packs” for the language in the area in question. In case of an emergency, task force members will be deployed with their devices, and when they arrive they can immediately determine exactly where resources need to be allocated by communicating with the locals. The emergency personnel–or the “interviewer”–initiates a session on the device and selects the preloaded language to begin. The interviewer and interviewee can then take turns communicating by speaking into the device, which immediately translates their words to the other user’s language. The interviewer can then ask the first critical questions, such as how many people are in the area, what resources are needed, or how the emergency task force can help. A text log of the conversation is retained on the Douglas and can be used later for further support.
The advantage of the Douglas over other translational devices is that it does not require Internet connection or cloud support–as programs such as Google Translate currently do–and can be used without any writing. The users can simply speak as if to a human translator and have their words be translated instantly. The program currently runs on an Android device and the proof-of-principle has been demonstrated in Cebuano, a language spoken in the Philippines by about 21 million people. Other intended languages for use include Malay, Indonesian, Ilocano, and Tok Pisin.
The program was initiated after Dr. Julie Vonwiller learned of the difficulties experienced by relief organisations–particularly RedR Australia–when communicating with the disaster victims in the field. Julie had previously worked with other organisations requiring translation in rapid time. Together with a colleague. Dr James Nealand, who was an expert in embedded systems, and supported by collaborators in Singapore and United States, they designed and built a proof of principle software system for Android, utilising off-the-shelf software systems where available.
An Android phone containing the Douglas app can be viewed at The Warren Centre for Advanced Engineering. For more information about the project, or questions about how to get involved, contact firstname.lastname@example.org