IEEE

Workshop on Speech Processing & Applications

Greetings from CMRIT.

We are pleased to announce a one-day Workshop on Speech Processing & Applications organized by the Dept. Electronics & Communication Engg., and the Dept. of Telecommunication Engg.  CMR Institute of Technology on June 21st 2012

Sponsors:  
CMR Institute of Technology, Bangalore, India
International Speech Communication Assn., Baixas, France

Technical Sponsor: 
IEEE Signal Processing Society, IEEE Bangalore section

Title: Monitoring Human Behavior: “Who Said What When and How?”

Speakers: Prof. John H. L. Hansen & Dr. Abhijeet Sangwan, UT, Dallas, Texas, USA

Venue: CMR Institute of Technology, 132, AECS Layout, IT Park Road, Bangalore 560 037, INDIA.
Date: Thursday, June 21, 2012 between 8:45 and 16:00 hrs

This workshop provides an overview of Modern Automatic Speech Recognition (ASR) systems’ in building products for efficacious personal services. Importantly, for faculties,  it helps in gaining knowledge about the vistas for research in Speech Processing and Applications.

Prof. John H. L. Hansen, UT, Dallas, Texas, USA
Abstract: There is significant interest in the development of effective human-machine interactive systems for a wider range of personal services. Speech and Speaker Processing/Recognition research has advanced significantly in recent years, but performance in real environments remains a major challenge. In this talk, we consider several recent research efforts in the field of robust speech and speaker recognition. The talk is partitioned into three phases:
(i) speech analysis/modeling,
(ii) monitoring speaker characteristics, and
(iii) applications – Prof-Life-Log and UTDrive In-Vehicle systems.
For Analysis & Modeling, challenges on the ability to identify those traits which distinguish an individual subject will be considered. Next, speaker monitoring including speaker stress/emotion, vocal effort (e.g., whisper, soft, neutral, loud, shout), Lombard effect (speech produced in noise), language/dialect, will be considered. Finally, applications of how to monitor human behavior will be explored. This will consider: (a) speech recognition and keyword spotting based on phonological features, (b) Speaker ID & Dialect/Language ID, and (c) Spoken Document Retrieval based on SpeechFind® (http://SpeechFind.utdallas.edu).

 

Dr. Abhijeet Sangwan, UT, Dallas, Texas, USA
Abstract: Small Vocabulary Speech Recognition  (Lecture and hands-on session)
Modern Automatic Speech Recognition (ASR) systems are complex machines. While research and development in ASR requires extensive expertise, building practical applications with existing state of the art can be relatively simpler and a rewarding experience. In this workshop, we will build a small vocabulary speech recognition engine to explore components of an ASR engine, namely,
(i) acoustic;
(ii) language; and
(iii) pronunciation models.
We will also point out a number of publicly available resources that can assist you in building practical ASR systems. The workshop can be immensely helpful for students who are working towards their projects and corporations that may be looking towards integrating speech recognition technology into their products and services. The workshop will use Sphinx-4, a state-of-the-art Java-based open source speaker independent continuous speech recognizer available under BSD style license. It provides a simple/powerful API (application programming interface) that allows you to rapidly build applications.

Profiles of Speakers:

John H. L. Hansen received his Ph.D. & M.S. degrees in Electrical Engineering from Georgia Institute of Technology, and B.S.E.E. degree from Rutgers Univ., College of Engineering, NJ. He joined Univ. of Texas at Dallas (UTD), Erik Jonsson School of Engineering & Computer Science in 2005, where he is Professor and Dept. Head of Electrical Engineering, and holds the Distinguished Univ. Chair in Telecommunications Engineering, and a joint appointment in School of Behavioral & Brain Sciences (Speech & Hearing). At UTD, he established The Center for Robust Speech Systems. He previously served as Dept. Chair and Prof. in Speech, Language and Hearing Sciences, and Prof. of Electrical & Computer Engineering, Univ. of Colorado at Boulder. He is an International Speech Communication Association (ISCA) Fellow, IEEE Fellow, Member and TC Chair of IEEE Signal Processing Society Speech & Language Processing Technical Committee, ISCA Distinguished Lecturer, and previously served as Member of IEEE, Signal Processing Educational Technical Committee, IEEE Signal Processing Society Distinguished Lecturer, Technical Advisor to U.S. Delegate for NATO (IST/TG-01), Associate Editor for IEEE Trans., Speech & Audio Processing, Associate Editor for IEEE Signal Processing Letters and Editorial Board Member for the IEEE Signal Processing Magazine. He has supervised 59 PhD/MS thesis candidates, was recipient of 2005 University of Colorado Teacher Recognition Award, and author/co-author of 429 journal and conference papers, including 10 books in the field of speech processing and language technology. He served as General Chair for Interspeech/ICSLP-2002: Inter. Conf. on Spoken Language Processing, and as Co-Organizer and Technical Program Chair for IEEE ICASSP-2010, Dallas, TX.

Abhijeet Sangwan received the Ph.D. and M.S. degrees in Electrical Engineering from UTD (2009) and Concordia University, Canada (2006), respectively. Currently, he is Assistant Research Professor at UTD and works on a number of federally funded research projects from DARPA (Defense Advanced Research Projects Agency), U.S. Air Force and NSF (National Science Foundation). His research interests include Automatic Speech Recognition (ASR), Automatic Language Identification, automatic accent assessment technologies, articulatory modeling, multimodal signal processing, acoustic environment estimation and characterization. He is also actively involved in industry academia partnership at UTD and has helped in successfully transferring key speech and language processing technology from the lab to industry. He is engaged with Speetra Inc. where he has helped in developing and delivering key speech recognition and speaking assessment technology. In 2008, he was part of the Human Language Technology group at The T.J. Watson Research Center, Yorktown Heights, New York where he worked on automatic sentiment analysis technology. He has published over 35 Journal and Conference papers, and holds 2 patents.

Looking forward to meet you all @ the workshop.

Thanks. Best regards.

*Prof. H. N. Shankar* https://sites.google.com/a/cmrit.ac.in/hnshankar/
Dean — Academics & Research
*CMR Institute of Technology*    http://it.cmr.ac.in/
132, AECS Layout, IT Park Road
Bangalore 560 037, INDIA
Tel: Dir: +91 (80) 6579 1751  ||  +91 (80) 2852 4630 &  2847 6965 x 333
Telefax : +91 (80) 2852 4477
hnshankar@cmrit.ac.in
hnshankarblr@gmail.com

Leave a Reply