BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//EE - ECPv5.10.0//NONSGML v1.0//EN
CALSCALE:GREGORIAN
METHOD:PUBLISH
X-WR-CALNAME:EE
X-ORIGINAL-URL:https://ee.iisc.ac.in
X-WR-CALDESC:Events for EE
BEGIN:VTIMEZONE
TZID:Asia/Kolkata
BEGIN:STANDARD
TZOFFSETFROM:+0530
TZOFFSETTO:+0530
TZNAME:IST
DTSTART:20240101T000000
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTART;TZID=Asia/Kolkata:20240129T140000
DTEND;TZID=Asia/Kolkata:20240201T153000
DTSTAMP:20260527T151107
CREATED:20240127T075025Z
LAST-MODIFIED:20240129T053328Z
UID:241314-1706536800-1706801400@ee.iisc.ac.in
SUMMARY:Lecture series on Electric Vehicles
DESCRIPTION:Dr Gautham Ram Chandra Mouli from TU Delft will visit us next week as part of an IISc-TU Delft collaborative effort. He will deliver a short (4-part) lecture series on electric vehicles from Jan 29th to Feb 1st from 2.00 pm to 3.30 pm.  \nThe venue will be Room B303 in the Department of EE. \nPlease sign up for the lecture series using the below link before 28/01/2024. \nhttps://forms.gle/8tQn5Mbaoh6s7tXdA
URL:https://ee.iisc.ac.in/event/lecture-series-on-electric-vehicles/
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Asia/Kolkata:20240206T113000
DTEND;TZID=Asia/Kolkata:20240206T130000
DTSTAMP:20260527T151107
CREATED:20240201T101134Z
LAST-MODIFIED:20240206T040039Z
UID:241319-1707219000-1707224400@ee.iisc.ac.in
SUMMARY:[EE Defense Talk]- Graph Based Approaches for Diarization of Conversational Speech\,
DESCRIPTION: The thesis defense talk of Ms. Prachi Singh (PhD candidate\, EE dept.) with the following details \nTitle Graph Clustering Approaches for Speaker Diarization of Conversational Speech \nDate and time  February 6\, 2024 (11:30am) \nVenue  MMCR\, EE\, (C241). And in Teams \nAbstract \nIn this era of advanced machine intelligence\, real-world speech applications still find it challenging to deal with conversations involving multiple speakers. An essential first step in speech information extraction from conversational speech is the task of finding “who spoke when”\, also referred to as speaker diarization. The focus of this talk is to describe our efforts in investigating graph representation learning and clustering techniques for this problem. While graph models have been used in several other domains\, our work on its application to temporal segmentation of speech is the first of its kind. \nThe talk is divided into three main parts. In the first part of this talk\, I will describe a novel proposal on self-supervised learning to perform joint representation learning and clustering\, called self-supervised clustering (SSC) for diarization. On the learned representations\, we explore path integral clustering (PIC)\, a graph-based clustering algorithm. The PIC is an unsupervised agglomerative graph clustering method that performs clustering based on the edge connections of a node\, called path integral. The proposed SSC with path integral clustering (SSC-PIC) is shown to achieve state-of-the-art performance for benchmark datasets. \nThe second part of the talk is an extension of SSC-PIC to incorporate metric learning. We design a neural version of the probabilistic linear discriminant analysis (PLDA) approach with learnable parameters to compute a log-likelihood score between embeddings from two segments of the recording.  We propose a joint self-supervised representation learning and metric learning approach called Selfsup-PLDA-PIC. \nIn the third part of the talk\, we develop a supervised learning setup using labeled conversational data for training. In this setting\, we propose a supervised clustering approach called Supervised HierArchical gRaph Clustering (SHARC) for speaker diarization. This approach uses Graph Neural Networks (GNN) to capture the similarity between the speaker embeddings and performs hierarchical clustering. An extension of this work is the joint training of the speaker embedding extractor along with the GNN module\, referred to as end-to-end SHARC (E-SHARC). I will also illustrate how to extend the E-SHARC model for diarization of overlapped speech recordings. \nThe talk will conclude with a summary of our key contributions\, highlighting the pros and cons of using graph-based models for speaker diarization. \n\n\n–—————–\nCoffee/tea will be served before the talk. All are welcome.
URL:https://ee.iisc.ac.in/event/ee-defense-talk-graph-based-approaches-for-diarization-of-conversational-speech/
LOCATION:Multi-Media Class Room (MMCR)\, EE Department (Hybrid mode)
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Asia/Kolkata:20240216T160000
DTEND;TZID=Asia/Kolkata:20240216T173000
DTSTAMP:20260527T151107
CREATED:20240207T040341Z
LAST-MODIFIED:20240207T040341Z
UID:241323-1708099200-1708104600@ee.iisc.ac.in
SUMMARY:EE Faculty Colloquium
DESCRIPTION:Title: Challenges & Opportunities in Dysarthric Speech ProcessingSpeaker: Dr. Prasanta Kumar Ghosh\, Dept of Electrical Engineering\, Indian Institute of ScienceVenue: MMCR\, EETime: 4pm\, Friday\, 16 FebAbstract:Dysarthria is a speech condition caused by motor impairments. This talk will give an overview of Dysarthric Speech Processing (DSP) arising due to Amyotrophic Lateral Sclerosis (ALS) and Parkinson’s Disease (PD). It will present our latest research results and bring out the challenges and opportunities in DSP.Speaker’s Bio:Prasanta Kumar Ghosh received his Ph.D. in Electrical Engineering from University of Southern California (USC)\, Los Angeles\, USA in 2011. Prior to that he obtained his M.Sc.(Engineering) in Electrical Communication Engineering from Indian Institute of Science (IISc)\, Bangalore and B.E.(ETCE) in Electronics from Jadavpur University\, Kolkata in 2006 and 2003 respectively. He has been a Research Intern at Microsoft Research India\, Bangalore in the area of audio-visual speaker verification from March to July in 2006. During 2011-2012 he was with IBM India Research Lab (IRL) as a researcher. Currently\, he is an assistant professor in the department of Electrical Engineering (EE) at IISc. Before joining as assistant professor\, he was a faculty fellow in the department of EE under the INSPIRE faculty fellowship program.Prasanta Kumar Ghosh was awarded the INSPIRE faculty fellowship from Department of Science and Technology (DST)\, Govt. of India in 2012. He was the winner of the first prize in Mr. BRV Varadhan Post-Graduate student paper contest in IEEE Bangalore chapter\, in 2005. He received the best M.Sc. (Engg.) thesis award for the year 2006-07 in the Electrical Sciences division at IISc. He was awarded Center of Excellence in Teaching’s award for excellence in teaching in the category of EE for the year 2010-11 in USC. He has also received the best teaching assistantship (TA) awards for the years 2007-08 and 2008-09 and the honorable mention for the best paper award in the EE\, USC. He was also awarded Ming Hsieh Institute (MHI) Ph.D. scholar for the year 2010-11 in EE\, USC. His research interests include non-linear signal processing methods for speech and audio\, speech production and its relation to speech perception\, and automatic speech recognition inspired by the speech production and perception link.————–
URL:https://ee.iisc.ac.in/event/ee-faculty-colloquium/
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Asia/Kolkata:20240224T090000
DTEND;TZID=Asia/Kolkata:20240224T170000
DTSTAMP:20260527T151107
CREATED:20240219T085818Z
LAST-MODIFIED:20240222T070820Z
UID:241328-1708765200-1708794000@ee.iisc.ac.in
SUMMARY:IISc EE Open Day 2024
DESCRIPTION:Click here for IISc EE Open Day \n  \n  \n  \n  \n 
URL:https://ee.iisc.ac.in/event/iisc-ee-open-day-2024/
LOCATION:IISc
END:VEVENT
END:VCALENDAR