Message Fly "Voice cloud"

Each cell phone will have the ability to speak;
Every household appliance will have the function of listening and speaking;
Each sedan will have the ability to talk;
Every toy will have the function of listening and speaking;
......
This is Anhui hkust Information Technology Co., Ltd. Senior vice President Wuxiaoyu in the CTI Forum reporter interview, for me to describe the voice of the vision, and in the previous years, voice applications, although the industry has been enough attention, but the real landing is undoubtedly the rapid development of mobile internet has a close relationship.

Photo: Mr. Wuxiaoyu, Senior vice president, Anhui hkust Information Technology Co., Ltd.

Message Fly "Voice cloud"

As Innovation Factory CEO Kai-Fu Lee in the Hkust news fly "voice lit life", "a new generation of voice cloud publishing and Voice Developers Conference" elaborated on: "In the past always said that voice changed the world, why has not changed?" On the one hand, cloud computing has not yet reached a high level. Second, voice in the past twenty or thirty years always do not know where to apply. The third challenge, which existed in the past, is that if a certain semantic understanding is achieved, it will take as little time as possible for developers to intervene quickly. The fourth challenge is the user experience and user expectations, voice is the most natural way to communicate with each other, so once people start to communicate with the voice, it means that people think of machines as a ' person ', such high expectations will give developers a greater challenge. ”

These challenges allow voice applications for a considerable period of time limited to some toys, car navigation and other fields, and let the sound through the barrier, free communication is the HKUST has been the dream of flying.

At the same time, in order to allow more Chinese, more Chinese enterprises to use the flying voice, in October 2010, Hkust launched the "Voice Cloud" developer program. Message flying "Voice cloud", is based on cloud computing technology, the industry-leading intelligent voice technology to the Internet developers open to all types of mobile internet entrepreneurs and innovative enterprises to provide low threshold of voice technology services, partners can be used as water, electricity, "that is, open, on demand", In a very short period of time to build a support for natural voice interaction features of mobile Internet applications.

and in less than 1.5 of the time, Wuxiaoyu said: "Our developer partners have reached 3,100, the number of end-users reached 30 million, the number of online users per day over 1.2 million, the total request for more than 7 million times, covering the mobile internet, gaming, automotive, communications, finance, school and other fields, Attract the active participation of engineers, it technicians and even students. In the mobile phone, call center, car, internet TV, smart home appliances and other fields have been innovative applications, ' voice cloud ' service has gone into thousands of households, daily life. "

Message to fly the next generation of" voice cloud "

Wuxiaoyu Introduction, compared to the first generation of" voice cloud ", the hkust flying a new generation of" voice cloud "platform is more robust, able to handle more complex business, and can serve more users, in addition, a new generation of voice cloud will open personalized recognition, Personalized synthesis, natural language understanding, voice print recognition, oral evaluation and other interfaces. The

Wuxiaoyu emphasizes: "The next generation of ' voice cloud ' has three main concerns, one is the increasing core voice technology: The new generation of voice cloud introduced a breakthrough voice technology, whether it is speech synthesis, speech recognition, voice transcription is currently the industry leader, only in 2011, Our patented technology is more than hundred. The second concern is the variety of voice technology, in addition to the speech synthesis, speech recognition and phonetic transcription techniques mentioned just now, we have added voice-print recognition, oral evaluation, natural language understanding technology, industry users can get more voice service types in the new voice cloud, and can make an organic combination according to their own characteristics. Expand the diversity of applications and services. The third concern is more personalized voice technology, can be targeted to solve the actual application of different user voice service unique, but also to meet the current mobile internet users personalized needs, so that voice interaction more vivid. "

Semantic Understanding, in human and human communication, mainly based on human interaction in the expression, gesture, intonation and other exchanges, to achieve the specific judgment of the semantic, and the Technical University of the interpretation of the understanding of the technology in the end how high recognition rate? The

Wuxiaoyu further explains: "Semantic understanding technology is still identifying certain well-defined areas of information that are specifically articulated." No matter at any time, voice technology can not do as people communicate with people, understand more deep-seated connotation. Hkust flying through the flat language, the application needs of a clear recognition of the semantic understanding of the higher rate. "

Flying voice in call center application

with the financial industry call Center on the application of more and more clear, compared to the traditional button menu, IVR Automatic service is becoming more and more verbose, cumbersome, complex, has seriously affected the user experience. and voice recognition Call center navigation technology allows users to directly say the demand, you can handle business, if you want to inquire account balance, just sayOut "Check balance" or "let me see how much money I have on my phone?" "The system can understand the user's intention and guide the user to transact the business."

Therefore, the Flying voice Recognition call center navigation technology has been used in ICBC telephone bank, and has been in Dalian, Xiamen, Qingdao, Ningbo, Lianyungang and other regions online. In addition, CITIC Bank telephone Bank has also been in Beijing, Shenzhen, Tianjin, Wuhan, Hefei, Guiyang, Harbin, Hohhot, Changchun and other regions online.

So, in each enterprise, industry, different business types, call center platform and business access is very different, voice access is very difficult? Wuxiaoyu pointed out: "At present, the Hkust voice products can be with the mainstream call center platform, such as Huawei, ZTE, Avaya, Edify, Genesys and other systems integration, can effectively support voice applications in these platforms, for different industries in different applications, Hkust flying set up a special project implementation team, to undertake the effect of optimization, user interaction design and other work to provide personalized services to ensure the success of the project. "

What is the future application prospect of speech technology in call centers? Wuxiaoyu gave a very positive answer: "Foreign call center applications of Intelligent Voice technology has been very common, the domestic key industries have also started to use the Hkust Intelligent voice technology, and has begun to pay off." Customers can take advantage of a new self-service voice service solution to address the growing need for information consulting, electronic transactions, and customer service, and to help users easily and naturally access information and services at all times and at any location through readily available telephones. "

Wuxiaoyu finally said:" The next 5 years, Hkust will continue to improve the voice technology level, to build more intelligent call center self-service channels to provide a steady stream of power.
 

(editor: Heritage)