Wang Xiaochuan Wuzhen show Sogou Si – Sohu is expected to replace the human machine technology-mine_清翼

Wang Xiaochuan Wuzhen show Sogou Si – Sohu is expected to replace the human machine technology Third World Internet Conference held in November 16th to 18 in Zhejiang, Wuzhen, mobile Internet forum, Sogou CEO Wang Xiaochuan and Facebook company vice president Shi Feng, a visiting professor at Stanford University, Jerry · Kaplan et al at the forum, and made the theme of "the future of artificial intelligence." speech. Wang Xiaochuan from the modern artificial intelligence technology "and not", and all the guests to the ultimate ideal, imagine the future of artificial intelligence he said: "the future of search is the era of artificial intelligence of the crown, search and input method to ask the future will move". It is worth mentioning that the forum site Wang Xiaochuan released a new product — Sogou artificial intelligence machine simultaneous real-time display of Machine Translation technology, the speech Chinese speech real-time speech recognition and simultaneous translation for guests English on screen display, caused a sensation. This is the world’s first real-time display Machine Translation technology of neural network in large activities based on the reliable and accurate rate is close to the human interpreter. Based on this technology, in addition to the voice of the scene on the screen in the speech recognition real-time display of Chinese content, as well as the content of the machine synchronous translation of English, real-time generation of the content of the bilingual subtitles rolling subtitles. From the live demonstration effect, Sogou voice recognition accuracy rate has exceeded 97%, Machine Translation 90% accuracy rate, the future is expected to replace human si. Simultaneous interpretation or history! Sogou since the research machine Si world debut internet conference following King Ogawa speech: Twelve speech in front of the guests, about technology, about the product, I hope to share some different content, as well as their own unique perspective. Today, we mentioned the AlphaGo, as the beginning of the initiation of artificial intelligence, in-depth learning in which the most important responsibility. Today when we began to think about, think the future of artificial intelligence can really replace people. So I would like to share more today to know what artificial intelligence can do today? What can’t be done? What is the ultimate ideal of the future? After AlphaGo, we see the most important breakthroughs in the field of speech and image, the progress in the field of text is very slow. Today, we have made some breakthroughs in Machine Translation, but the question and answer and semantic understanding is not enough. Back to the Turing test, the last century 50s Turing machine that ask such a concept, we feel today is the voice of image progress quickly, but natural language processing is relatively slow. Regardless of the technique, as a product manager, artificial intelligence has three product directions, one is recognition — speech recognition, image recognition, video image recognition; two – we are to produce images, generate recognition; the three is to create. We talked about the level of artificial intelligence, I would like to change a way to describe the Engineer in the era of artificial intelligence will be in an increasingly important position. We began to refer to the traditional approach is to teach the rules to the machine, with the development of statistical systems, including deep learning!相关的主题文章: