IBM is donating speech translation technology to the United States government to break down language barriers and support better communication in Iraq. IBM will be sending 1,000 devices, which are code-named MASTOR (for Multilingual Automatic Speech Translator), and providing 10,000 copies of the software for the government’s future use (the donation is worth $45 Millions USD). These sophisticated, two-way “speech-to-speech” translators will help English and Iraqi Arabic speakers understand one another better, addressing a major concern among military personnel, their families and civilians in Iraq. MASTOR has been tested to run on a variety of platforms, including PDA, tablet PC and laptop computers.
IBM has a long history of providing technology and services to assist the humanitarian efforts of governments and organizations around the world. With its vast size and global reach, the company and the communities in which it operates are often directly affected by any event. In this case, there are 160 IBMers on active duty in Iraq and Afghanistan. It was these employees, and most recently Sgt. Mark Ecker, the son of a long-time IBMer, who provided the impetus for IBM to get involved to help improve the safety of U.S. and coalition personnel, as well as citizens and aid organizations. More after the jump!
â€œIBM employees returning from service with the U.S. military in Iraq have consistently emphasized two points: the importance of communicating with the Iraqi people and the operational challenges posed by the need to do so,” said Sam Palmisano, IBM Chairman and CEO. “Although in many instances human translators are essential, we also believe that there are technological solutions to help mitigate the problem.”
IBM began developing the MASTOR technology at IBM’s T.J. Watson Research center in 2001 and gained development support as part of the DARPA (Defense Advanced Research Project Agency) Spoken Language Communication and Translation System for Tactical Use (TRANSTAC) program. According to Michael Picheny, senior manager for speech and language algorithms, the support of DARPA has been critical. â€œThey set the overall framework for our efforts, provide a set of shared resources, and give critical feedback to help us improve our technology.â€ In October 2006, IBM successfully delivered MASTOR to the U.S. Joint Forces Command for field testing for potential deployment in Iraq. “Such field testing is invaluable in helping us harden the system for actual usage and we are grateful to JFCOM for their confidence.” said Picheny.
While much work remains to be done in speech and translation technologies, MASTOR is at the point where it can provide real value.
By easing communications, MASTOR can help U.S. and coalition personnel distributeessentials, such as food, clean water, clothing, electricity and medical treatment. Yuqing Gao, manager for speech recognition and understanding, says, â€œIf military personnel can communicate with local people, it will eliminate unnecessary conflicts caused by language barriers. If this technology can help people communicate better, then we could have a better world.”
What makes MASTOR important…and different?
There are fewer than 20 commercial translation systems currently available globally. Yet the need for cross-language communication has never been more urgent. A secondary goal of IBMâ€™s contribution is to encourage other private sector organizations to speed their translation development and deployment, advance collaboration among this community of innovators, and prompt additional companies to extend their resources for similar humanitarian missions.
Unfortunately, many of the available commercial translation systems can work only with preprogrammed fixed phrases â€“ significantly limiting a two-way bilingual conversation. By combining automatic speech recognition, natural language understanding and speech synthesis technologies, IBMâ€™s MASTOR translation technology offers users the ability to converse naturally without having to memorize any predetermined phrases. This results in the ability to convey meaning even if the system makes minor errors. MASTOR can recognize and translate a vocabulary of over 50,000 English and 100,000 Iraqi Arabic words. The technology is also available in Mandarin Chinese and IBM is currently extending it to additional languages.
To operate, the user speaks into a microphone linked with MASTOR. The technology recognizes and translates the speech, then vocalizes the translation in the target language. Upon hearing his or her language spoken, the person can respond in his or her native language and MASTOR will translate it back into the original language. MASTOR produces an audible and text translation of the spoken words that can run on a variety of devices, including PDAs, tablet PCs or laptop computers.
With its ability to be portable and flexible, MASTOR could ultimately add significant business value in many different industries.
The upcoming Beijing Olympics, for example, presents an opportunity for the government officials in the Chinese public sector to provide their bus drivers with tools to make public transportation easier for its visitors. And Gao suggests that “businesses with global knowledge workers could use this tool for conference calls or video conferences, which would be much easier if employees could speak in their native languages. For companies with stores around the world, this would help their businesses grow faster and easier. And as India’s economy matures and grows, other parts of the world could use MASTOR to gain an edge for the call center jobs. It could also help with the shortage of doctors in third-world nations as they wouldn’t have to learn the local language before going, which could open up whole new fields of opportunity.”