BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:Probabilistic Approaches to Robot Task Benchmarking
DTSTART:20160831T150000
DTSTAMP:20260507T174151Z
UID:7fea3e29ca5675adfd8208de7cf6a1bfa0fab60580e3009faa62a5d8
CATEGORIES:Conferences - Seminars
DESCRIPTION:Dr. Mathew Doss\, IDIAP\, Martigny\nOne of the crucial steps i
 n development of speech technologies is to train statistical models that c
 apture the relationship between the observed speech signal and the classes
  representing the high level information that we are interested in. Tradit
 ionally\, development of such statistical models is divided into two steps
 \, namely\,\n(a) extraction of "hand-crafted" features using signal proces
 sing techniques and prior knowledge about speech production and perception
 \, and\n(b) training a classifier with the extracted features as input. Fo
 r example\, in automatic speech recognition systems typically short-term s
 pectrum based features are first extracted\, and then subsequently modeled
  by Gaussian mixture models or artificial neural networks to estimate phon
 e class likelihoods or posterior probabilities.\nIn this talk\, I will pre
 sent a novel approach\, originally developed at Idiap\, where the relevant
  features and the classifier are jointly learned from the raw speech signa
 l using convolutional neural networks (CNNs). The talk will demonstrate th
 e potential of the proposed approach through two different speech processi
 ng studies\, namely\, automatic speech recognition study and anti-spoofing
  study (in the context of automatic speaker verification). Specifically\, 
 I will show how\, with minimal prior knowledge or assumptions\, the propos
 ed CNN-based approach learns to transform the speech signal and model the 
 relevant information to yield better systems.\nBio: Dr. Mathew Magimai Dos
 s received the Bachelor of Engineering (B.E.) in Instrumentation and Contr
 ol Engineering from the University of Madras\, India in 1996\; the Master 
 of Science (M.S.) by Research in Computer Science and Engineering from the
  Indian Institute of Technology\, Madras\, India in 1999\; the PreDoctoral
  diploma and the Docteur dès Sciences (Ph.D.) from Ecole Polytechnique F
 édérale de Lausanne (EPFL)\, Switzerland in 2000 and 2005\, respectively
 . He was a postdoctoral fellow at International Computer Science Institute
  (ICSI)\, Berkeley\, USA from April 2006 till March 2007.\nSince April 200
 7\, he has been working as a Researcher in the Speech and Audio Processing
  group at Idiap Research Institute\, Martigny\, Switzerland. He is also a 
 lecturer at EPFL. He is an associate editor of the IEEE Signal Processing 
 Letters. His current research interests include signal processing\, statis
 tical pattern recognition\, artificial neural networks and computational l
 inguistics with applications to automatic speech recognition\, automatic s
 peaker recognition\, objective speech assessment\, spoken language process
 ing and automatic sign language recognition and assessment.
LOCATION:MEB110 http://plan.epfl.ch/?request_locale=fr&room=meb110&domain=
 places
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR