{"id":266310,"date":"2014-05-27T17:58:12","date_gmt":"2014-05-28T00:58:12","guid":{"rendered":"https:\/\/www.microsoft.com\/en-us\/research\/?p=266310"},"modified":"2016-07-26T15:10:07","modified_gmt":"2016-07-26T22:10:07","slug":"enabling-cross-lingual-conversations-real-time","status":"publish","type":"post","link":"https:\/\/www.microsoft.com\/en-us\/research\/blog\/enabling-cross-lingual-conversations-real-time\/","title":{"rendered":"Enabling Cross-Lingual Conversations in Real Time"},"content":{"rendered":"
Sometimes, a dream needs but a nudge to propel it toward reality.<\/p>\n
That is exactly what Eric Rudder<\/a> and Peter Lee<\/a> delivered to Microsoft\u2019s Machine Translation<\/a> team about a year ago.<\/p>\n Rudder, then Microsoft\u2019s chief research and strategy officer, and Lee, head of Microsoft Research, had been among those who had seen demonstrations of a research project designed to provide fluent, cross-lingual conversations between speakers of different languages. The demos were tantalizing, offering the potential to smash through the language barriers that have hindered human communication for thousands of years.<\/p>\n Rudder and Lee were convinced: The time had come to make it real.<\/p>\n That task fell to, among others, the Machine Translation team, based in Redmond and headed by Arul Menezes<\/a>.<\/p>\n The success of the team\u2019s progress to date was on display May 27, in a talk by Microsoft CEO Satya Nadella<\/a> in Rancho Palos Verdes, Calif., during the Code Conference<\/a>. During Nadella\u2019s conversation with Kara Swisher and Walt Mossberg of the Re\/code tech website relating to a new era of personal computing, he asked Gurdeep Pall to join him on stage. Pall, the Microsoft corporate vice president of Skype<\/a>, demonstrated for the first time publicly<\/a> the Skype Translator app, with Pall conversing in English with German-speaking Microsoft employee Diana Heinrichs.<\/p>\n The interaction underscored the steady advancement the team has achieved.<\/p>\n \u201cWe felt speech translation was a very natural evolution of the text-translation work we\u2019ve been doing,\u201d says Chris Wendt<\/a>, program manager of the Machine Translation team, who had long been pushing for such an incubation effort.<\/p>\n \u201cIt\u2019s an exciting project,\u201d he says, \u201cand it became clear that adding this capability to Skype and enabling people to have translated conversations was the killer scenario to get this technology into customers\u2019 hands.\u201d<\/p>\n There have been many attempts over the years, several within Microsoft Research, to demonstrate such aspects of translating human speech. But delivering something that is usable in real life, to fit the voice and utterances of many different users and the nuances of different languages\u2014all of it built at scale to serve Skype users\u2014has been considered a nearly impossible task.<\/p>\n Making Skype Translator available first on Windows 8<\/a> later this year as a limited beta has required a series of remarkable research advances in translation, speech recognition, and language processing, combined with contributions from Microsoft engineering and research teams near and far.<\/p>\n The Machine Translation team, already responsible for the state-of-the-art translation technology and the cloud service used extensively in a number of Microsoft products, has taken a One Microsoft approach to the challenge by utilizing contributions from researchers and engineers working on Microsoft\u2019s speech service. Additionally, to develop a new, yet natural user experience, the team developed a deep partnership with Skype\u2019s designers and engineers, particularly the prototyping team led by Jonas Lindblom.<\/p>\n \u201cThe interesting thing about this project,\u201d Menezes says, \u201cis we\u2019ve got these two fairly complex technologies coming together for the first time to provide this end-to-end user experience.\u201d<\/p>\n Microsoft Research has been focused on machine translation for more than 10 years. Initial results came with translations for Microsoft\u2019s product-support Knowledge Base. The technology became available for public use as the engine behind Bing Translator<\/a>, spurring the tool\u2019s growth and popularity since its launch. Today, Bing Translator is used by millions and embraced by organizations worldwide.<\/p>\n Along the way, Menezes and Microsoft colleagues have addressed significant system and user-interface design challenges, including reducing latency and developing visual feedback so the translation system is continuously improving itself using user feedback.<\/p>\n \u201cThe technology is only as good as the data,\u201d Menezes says. \u201cOne big focus has been to scale up the amount and kinds of data that go into the machine-learning training of these systems.\u201d<\/p>\n