BEGIN:VCALENDAR VERSION:2.0 PRODID:IEEE vTools.Events//EN CALSCALE:GREGORIAN BEGIN:VTIMEZONE TZID:Asia/Kolkata BEGIN:STANDARD DTSTART:19451014T230000 TZOFFSETFROM:+0630 TZOFFSETTO:+0530 TZNAME:IST END:STANDARD END:VTIMEZONE BEGIN:VEVENT DTSTAMP:20230906T104128Z UID:5060C5BF-5B48-4558-8034-C00160E26EF0 DTSTART;TZID=Asia/Kolkata:20230824T100000 DTEND;TZID=Asia/Kolkata:20230824T110000 DESCRIPTION:Around half of the information humans exchange during interacti on is not the meaning of speech itself. Speech audio signal carries inform ation about the age\, gender\, emotion of the speaker. In this talk we wil l discuss the information that can be obtained from the speech signal\, po tential approaches and applications. Then we will extend the scope with ex tracting information from non-speech audio – audio events detection and audio background recognition. We will discuss the technologies and algorit hms for solving these problems – neural networks with supervised and non -supervised training\, most commonly used features and cost functions.\n\n Speaker(s): Dr. Ivan Tashev \, \n\nVirtual: https://events.vtools.ieee.org /m/370590 LOCATION:Virtual: https://events.vtools.ieee.org/m/370590 ORGANIZER:ieee.sps.sb.iitkgp@gmail.com SEQUENCE:23 SUMMARY:Audio Analytics – What We Can Get from Speech Beyond Speech Recog nition\, and is There Anything Useful in the Non-Speech Audio URL;VALUE=URI:https://events.vtools.ieee.org/m/370590 X-ALT-DESC:Description: <br /><p><span style="font-weight: 400\;">Around ha lf of the information humans exchange during interaction is not the meanin g of speech itself. Speech audio signal carries information about the age\ , gender\, emotion of the speaker. In this talk we will discuss the inform ation that can be obtained from the speech signal\, potential approaches a nd applications. Then we will extend the scope with extracting information from non-speech audio &ndash\; audio events detection and audio backgroun d recognition. We will discuss the technologies and algorithms for solving these problems &ndash\; neural networks with supervised and non-supervise d training\, most commonly used features and cost functions.&nbsp\;</span> </p> END:VEVENT END:VCALENDAR