{"id":585154,"date":"2019-05-09T13:13:15","date_gmt":"2019-05-09T20:13:15","guid":{"rendered":"https:\/\/www.microsoft.com\/en-us\/research\/?post_type=msr-project&p=585154"},"modified":"2020-11-12T13:43:43","modified_gmt":"2020-11-12T21:43:43","slug":"project-denmark","status":"publish","type":"msr-project","link":"https:\/\/www.microsoft.com\/en-us\/research\/project\/project-denmark\/","title":{"rendered":"Project Denmark"},"content":{"rendered":"

The goal of Project Denmark is to move beyond the need for traditional microphone arrays, such as those supported by Microsoft\u2019s Speech Devices SDK<\/a>, to achieve high-quality capture of meeting conversations. Instead, virtual microphone arrays are dynamically created from consumer-grade devices, such as mobile phones or laptops equipped with ordinary microphones. \u00a0Algorithms for combining speech information at multiple levels yield transcription accuracy that approaches that from close-talking microphones, with many potential applications. For example, we may pair multiple Microsoft Translator<\/a> applications to help meeting participants communicate more effectively across language barriers. Furthermore, meeting understanding<\/a> technology may be enabled by the speaker-attributed transcripts resulting from Project Denmark, without specially designed hardware<\/a>.<\/p>\n\t\t\t

\n\t\t\t
\n\t\t\t\t\t
\n\t\t