By Jia Wu, China Internet Weekly
December 20, 2009 12:00 PM PT
Jackson receives an e-mailed diagnostic report from the hospital, in which "refrain from using voice" is highlighted in red. He has recently been plagued by severe laryngitis, but a meeting with one of his important clients is awaiting him in the afternoon. Bearing this in mind, Jackson switches on his computer and loads his well-prepared script in the speech synthesis system.
3:00 pm, the same day. Jackson’s convincing voice is falling gently on all the ears in the session room, none of them detecting anything unusual. All of the client’s questions are met by fluent answers, but the voice isn’t coming from Jackson's throat. Instead, they are coming from a “virtual announcer” installed on a PC — all Jackson needs to do is type in the words he wants to speak.
"It was ten years ago that we started the virtual announcer project, and the process of improving its algorithm could have been developed into a work of science fiction," said Eric Chang, Director of Technology Strategy at Microsoft Research Asia with a smile, "It is still expensive to use virtual announcers, but we hope that in the next ten years, interaction and communication will be increasingly smooth and that peripherals will be much easier to apply. I personally have been looking forward to such a nice surprise."
Things tend to operate on a ten-year cycle. Just having celebrated its 10th birthday, Microsoft Research Asia is already planning development for the coming 10 years. According to Chang, the next decade will see greater availability of knowledge and wisdom accumulated online.
The world before "Star Wars" was one with no mobile phones and personal computers, let alone the "Internet." Human beings were making their first few steps into outer space. Powered by his infinite imagination, George Lucas laid his hands on the most advanced digital technologies at that time to present incredible visions of space. Apart from its spectacular box office revenue, Lucas’ creative film also marked the beginning of the digital age in cinematography.
"While talking about my job, I jokingly borrow the idea of C3PO, a character in ‘Star Wars,’" Chang told China Internet Weekly, adding, "the C3 refers to ‘connection, communication, and collaboration.’ The P stands for ‘patent,’ as all researchers here hope to achieve something, so we invest heavily in patent applications. The O is short for ‘others.’"
Projects at Microsoft Research Asia are like voyages in "Star Wars," where the spirit of adventure consistently drives the episodes.
During U.S. President Barack Obama's visit to China, Craig Mundie, Presidential Technology Advisor and Chief Research and Strategy Officer of Microsoft Corporation made a special trip to Microsoft Research Asia. "He came to remind us that our job is to take risks," Chang said, "Any researcher aspiring for technical innovation must identify problems or challenges that are significant enough to support long-term projects that will last for three to five years or longer. A simple project that anyone would know how to handle at a glance is not suited to a research lab."
"If every project is successful, the lab itself is a failure. That would mean that we are not taking enough risks."
Adventure is not audacity. For Microsoft, an enterprise that "refuses to lag behind," all adventures should lead to a substantial impact on future technology or lifestyles. "A more effective approach to this great goal is to encourage everyone to pursue the same dream with you. Accomplishing things that involve high risks usually relies on cooperation among people with varying specialties who can together discover methods with a better chance for success." Inter-group cooperation has been a leading research mode at Microsoft Research Asia in 2009.
Arthur Charles Clarke, a well-known science fiction writer, once said that “any sufficiently advanced technology begins to resemble magic.” You see it in action but remain clueless about its mechanism. Such technology is what Microsoft Research Asia focuses on.
"Our work is increasingly related to that of the outer world." Speaking of inter-group cooperation and cooperation between Microsoft Research Asia and the Microsoft Product Team, Chang cited the “Virtual Announcer” and Engkoo as his examples. "The 'Virtual Announcer' was completed through a multilateral cooperation led by the Speech Group, with powerful support from Visual Computing Group. In addition to voice processing, you have to deal with changes in various facial features and then the convergence,” Chang said. “A large dose of cooperation is needed." Engkoo, for its part, made its way out of the four walls of Microsoft Research Asia and showed up on MSN pages. "We worked with the Product Team so that more people would come to know it, use it, and provide feedback. Then we made further improvements based on the feedback."
The mode of research where preliminary findings are placed on the website to solicit comments from users is called Deployment Driven Research at Microsoft Research Asia. As technologies are developing at a speed beyond anyone’s imagination, computer science is also winding its way into different application scenarios. "In the past, computer technologies were often used to address well defined problems, but today the problems are getting more and more mixed, in that their solutions are impossible if you do not know how ordinary people would use such technologies."
In fact, Chinese people who need to use the English language outnumber the population of the U.S. but language tools currently available are not good enough to meet such demand. Comprehensive and in-depth communication with users is indispensable if one wants to understand people’s actual needs, such as particularly handy functions or the most frequently used features. “An advantage of ‘Deployment Driven Research’ lies in the possibility to understand users’ needs in a collective manner. Engkoo, for instance, is designed for white-collar workers and college students, but some users demand a differentiated interface for students in primary schools or junior high schools who prefer an interface with rich colors. We often fail to think of these ideas but they reflect the needs of users," Chang said.
Through interactions with users in the real world, Microsoft Research Asia has managed to better understand if certain technology should be included in a given product, or to decide an appropriate time for such productization. New technologies are being created and upgraded thanks to constant communication with users.
Quietly, the Internet is building connections among individuals, machines, or even between individuals and machines. The world of tomorrow may exceed the wildest imaginations of today.
On November 3, Chinese Premier Wen Jiabao delivered a speech entitled "Sciences and Technologies Shall Power Sustainable Development in China" to the academic community in Beijing, in which he pointed out that the information industry is the major driving force behind global economic recovery.
Social networking websites are digitizing people’s daily interactions, and digitalized protection of cultural heritage is already a normal practice. Digital healthcare and smart railways are effectively collecting, storing, analyzing and applying massive amounts of data. "There is going to be stronger demand for large-scale data utilization," said Chang, adding after a slight pause, "Relations between 'clouds' and 'clients' are getting closer. For instance, how would you synchronize or actively process data on separate devices? If we shift to a broader model of thinking, 'three screens plus a cloud' (PC, mobile phone, and TV + cloud computing) may be the future you want."
Wherever you are, as long as you have any of the most portable terminals, you can have access to information you desire, and automatically and smoothly transfer this data whenever you lay your hands on a larger screen. "We still have a long way to go in providing better and more effective means for interpersonal communication."
Toward the end of the interview, Chang told China Internet Weekly that the uniqueness of Microsoft Research Asia lies in the fact that outstanding researchers are allowed to pursue whatever projects they desire. More importantly, researchers may also find opportunities here to exert and amplify their influence by converting their research findings into marketable products, for example. More often than not there is where Microsoft’s rhapsody about the future begins.