FACULTY RESEARCH FELLOW PROGRAMME - 2015

Ms. Pankti Brar, Assistant Professor, Computer Science Engineering Department of Dronacharya College of Engineering, Gurgaon attended a two month Summer Faculty Research Fellow Program - 2015 from 15 May 2015 to 15 July 2015 at IIT Delhi.

The summer faculty research fellowship (SFRF) program is conducted by IIT Delhi to inculcate and enhance the research culture in Faculties of Engineering colleges. It was great opportunity for Ms. Pankti Brar to work under the mentorship of Prof./Dr. Arun Kumar, HOD, Centre of Applied research in Electronics (CARE) on “Speech signal processing and Artificial Neural Networks”.

Speech signals are actually produced from the air coming out of the lungs which is filtered by the vocal tract and various sounds are given out because of the effect of articulatory organs like the velum, lips, tongue etc. The speech signal has various regions that are voiced, unvoiced and silence regions. The voiced regions are the ones that are produced by the vibrations of the vocal chord. The short term energy of these signals are highest amongst the three. The unvoiced regions in the speech waveform are when the vocal cords don’t vibrate, these have a very low energy. The silence region is the one in which no speech signal is produced and this portion has a very minimal energy or zero energy.

Processing of speech signal using Artificial Neural networks is a very interesting and upcoming field. This processing involves speech recognition and speech synthesis. The Bidirectional Recurrent Neural Network (BRNN) with Long Short term Memory (LSTM) have proved out to be very revolutionary in this field.

This program was very beneficial for the faculties because it helped to realize the core interest area of research for the participants. It provided an opportunity to see the practical implementation of many important aspects that are usually studied theoretically. It also updated participants on the current research work that is being done in the area of acoustic signal processing, the problems that are being faced in speech recognition and speech synthesis with the current technologies and the new features that are being adopted to overcome those difficulties. The entire duration of the program was very knowledge enhancing and enriching for professional as well as personal growth.