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:20260204T182826Z
UID:FDD52096-0C4E-421F-AB77-A2683E23976C
DTSTART;TZID=Asia/Kolkata:20240618T123000
DTEND;TZID=Asia/Kolkata:20240618T133000
DESCRIPTION:With the emergence of advanced machine learning techniques and 
 large-scale datasets\, holistic analysis of realistic soundscapes becomes 
 more and more appealing. In the case of everyday soundscapes this can mean
  not only recognizing what are the sounds present in an acoustic scene\, b
 ut also where they are located and when they occur. This talk will discuss
  the task of joint detection and localization of sound events addressing t
 he above problem. The state of the art methods typically use spectral repr
 esentations and deep neural networks based on convolutional\, recurrent\, 
 and attention layers that share many similarities to neighboring fields. H
 owever\, the task also has several unique challenges\, which will require 
 specific solutions. We will give an overview of the task setup for trainin
 g machine learning models\, acoustic features for representing multichanne
 l signals\, topologies of deep neural networks\, and loss functions for tr
 aining systems. Since the performance of the methods is heavily based on t
 he training data used\, we will also discuss datasets that can be used for
  the development of methods and their preparation. We will discuss the rec
 ent DCASE evaluation campaign tasks that addressed the problem of joint de
 tection and localization of sound events.\n\nSpeaker(s): Dr. Tuomas Virtan
 en\, \n\nVirtual: https://events.vtools.ieee.org/m/424604
LOCATION:Virtual: https://events.vtools.ieee.org/m/424604
ORGANIZER:ieee.sps.sb.iitkgp@gmail.com
SEQUENCE:16
SUMMARY:IEEE SPS SBC Webinar: Detection and Localization of Sound Events (B
 y Dr. Tuomas Virtanen)
URL;VALUE=URI:https://events.vtools.ieee.org/m/424604
X-ALT-DESC:Description: <br /><p dir="ltr">With the emergence of advanced m
 achine learning techniques and large-scale datasets\, holistic analysis of
  realistic soundscapes becomes more and more appealing. In the case of eve
 ryday soundscapes this can mean not only recognizing what are the sounds p
 resent in an acoustic scene\, but also where they are located and when the
 y occur. This talk will discuss the task of joint detection and localizati
 on of sound events addressing the above problem. The state of the art meth
 ods typically use spectral representations and deep neural networks based 
 on convolutional\, recurrent\, and attention layers that share many simila
 rities to neighboring fields. However\, the task also has several unique c
 hallenges\, which will require specific solutions. We will give an overvie
 w of the task setup for training machine learning models\, acoustic featur
 es for representing multichannel signals\, topologies of deep neural netwo
 rks\, and loss functions for training systems. Since the performance of th
 e methods is heavily based on the training data used\, we will also discus
 s datasets that can be used for the development of methods and their prepa
 ration. We will discuss the recent DCASE evaluation campaign tasks that ad
 dressed the problem of joint detection and localization of sound events.</
 p>\n<p>&nbsp\;</p>
END:VEVENT
END:VCALENDAR