{"582393":{"#nid":"582393","#data":{"type":"event","title":"IEEE SP Talk - Machine Learning Approaches for Speech Enhancement","body":[{"value":"\u003Cp class=\u0022p1\u0022\u003E\u003Cspan style=\u0022line-height: 1.6em;\u0022\u003EAbstract:\u0026nbsp;During the last few years, machine learning has started\u003C\/span\u003E\u003Cspan style=\u0022line-height: 1.6em;\u0022\u003E\u0026nbsp;to permeate the world of speech enhancement and has produced\u003C\/span\u003E\u003Cspan style=\u0022line-height: 1.6em;\u0022\u003E\u0026nbsp;results that drastically improve over the state of the art. In\u003C\/span\u003E\u003Cspan style=\u0022line-height: 1.6em;\u0022\u003E\u0026nbsp;this talk I\u0026rsquo;ll touch on some of the most recent approaches on\u003C\/span\u003E\u003Cspan style=\u0022line-height: 1.6em;\u0022\u003E\u0026nbsp;both multichannel and single channel enhancement, and I will show\u003C\/span\u003E\u003Cspan style=\u0022line-height: 1.6em;\u0022\u003E\u0026nbsp;how traditional signal processing approaches can be reimagined\u003C\/span\u003E\u003Cspan style=\u0022line-height: 1.6em;\u0022\u003E\u0026nbsp;using machine learning tools such as mixture models, matrix\u003C\/span\u003E\u003Cspan style=\u0022line-height: 1.6em;\u0022\u003E\u0026nbsp;factorizations, deep learning regressions, and more.\u003C\/span\u003E\u003Cspan style=\u0022line-height: 1.6em;\u0022\u003E\u0026nbsp;\u003C\/span\u003E\u003C\/p\u003E\r\n\r\n\u003Cp class=\u0022p1\u0022\u003E\u003Cspan style=\u0022line-height: 1.6em;\u0022\u003EBiography:\u0026nbsp;Paris Smaragdis is an associate professor at the\u0026nbsp;\u003C\/span\u003E\u003Cspan style=\u0022line-height: 1.6em;\u0022\u003EComputer Science and the Electrical and Computer Engineering\u003C\/span\u003E\u003Cspan style=\u0022line-height: 1.6em;\u0022\u003E\u0026nbsp;departments of the University of Illinois at Urbana-Champaign, as\u003C\/span\u003E\u003Cspan style=\u0022line-height: 1.6em;\u0022\u003E\u0026nbsp;well as a senior research scientist at Adobe Research. He\u003C\/span\u003E\u003Cspan style=\u0022line-height: 1.6em;\u0022\u003E\u0026nbsp;completed his masters, PhD, and postdoctoral studies at MIT,\u003C\/span\u003E\u003Cspan style=\u0022line-height: 1.6em;\u0022\u003E\u0026nbsp;performing research on computational audition. In 2006 he was\u003C\/span\u003E\u003Cspan style=\u0022line-height: 1.6em;\u0022\u003E\u0026nbsp;selected by MIT\u0026rsquo;s Technology Review as one of the year\u0026rsquo;s top\u003C\/span\u003E\u003Cspan style=\u0022line-height: 1.6em;\u0022\u003E\u0026nbsp;young technology innovators for his work on machine listening, in\u0026nbsp;\u003C\/span\u003E\u003Cspan style=\u0022line-height: 1.6em;\u0022\u003E2015 he was elevated to an IEEE Fellow for contributions in audio\u003C\/span\u003E\u003Cspan style=\u0022line-height: 1.6em;\u0022\u003E\u0026nbsp;source separation and audio processing, and during 2016-2017 he\u003C\/span\u003E\u003Cspan style=\u0022line-height: 1.6em;\u0022\u003E\u0026nbsp;is an IEEE Signal Processing Society \u0026nbsp;Distinguished Lecturer. He\u003C\/span\u003E\u003Cspan style=\u0022line-height: 1.6em;\u0022\u003E\u0026nbsp;has authored more than 100 papers on various aspects of audio\u003C\/span\u003E\u003Cspan style=\u0022line-height: 1.6em;\u0022\u003E\u0026nbsp;signal processing, holds more than 40 patents worldwide, and his\u003C\/span\u003E\u003Cspan style=\u0022line-height: 1.6em;\u0022\u003E\u0026nbsp;research has been productized by multiple companies.\u003C\/span\u003E\u003C\/p\u003E\r\n","summary":null,"format":"limited_html"}],"field_subtitle":"","field_summary":[{"value":"\u003Cp\u003E\u003Cspan style=\u0022line-height: 20.8px;\u0022\u003EIEEE Signal Processing Society Distinguished Lecture\u003C\/span\u003E\u003C\/p\u003E\r\n","format":"limited_html"}],"field_summary_sentence":[{"value":"IEEE Signal Processing Society Distinguished Lecture"}],"uid":"27469","created_gmt":"2016-10-11 16:43:52","changed_gmt":"2017-04-13 21:14:22","author":"Kristen Bailey","boilerplate_text":"","field_publication":"","field_article_url":"","field_event_time":{"event_time_start":"2016-10-25T12:45:00-04:00","event_time_end":"2016-10-25T14:00:00-04:00","event_time_end_last":"2016-10-25T14:00:00-04:00","gmt_time_start":"2016-10-25 16:45:00","gmt_time_end":"2016-10-25 18:00:00","gmt_time_end_last":"2016-10-25 18:00:00","rrule":null,"timezone":"America\/New_York"},"extras":[],"groups":[],"categories":[],"keywords":[],"core_research_areas":[],"news_room_topics":[],"event_categories":[{"id":"1795","name":"Seminar\/Lecture\/Colloquium"}],"invited_audience":[{"id":"78761","name":"Faculty\/Staff"},{"id":"78771","name":"Public"},{"id":"78751","name":"Undergraduate students"},{"id":"174045","name":"Graduate students"}],"affiliations":[],"classification":[],"areas_of_expertise":[],"news_and_recent_appearances":[],"phone":[],"contact":[{"value":"\u003Cp class=\u0022p1\u0022 style=\u0022line-height: 20.8px;\u0022\u003E\u003Cspan class=\u0022s1\u0022\u003EAlessio Medda\u003C\/span\u003E\u003C\/p\u003E\r\n\r\n\u003Cp class=\u0022p2\u0022 style=\u0022line-height: 20.8px;\u0022\u003E\u003Cspan class=\u0022s2\u0022\u003E\u003Ca href=\u0022mailto:alessio.medda@gtri.gatech.edu\u0022\u003E\u003Cspan class=\u0022s3\u0022\u003Ealessio.medda@gtri.gatech.edu\u003C\/span\u003E\u003C\/a\u003E\u003C\/span\u003E\u003C\/p\u003E\r\n","format":"limited_html"}],"email":[],"slides":[],"orientation":[],"userdata":""}}}