About me
I am currently working as a Postdoc with
Dr. Petr Motlicek at the
IDIAP Research Institute in Switzerland. My work here is to develop ASR solutions for Air Traffic Control communications. Previously, I was engaged as a Senior Project Engineer at
IIT Dharwad in Karnataka, India in a project for building speaker recognition systems for Indian languages. The project was sponsored by the
National Language Translation Mission of MeitY, Govt. of India. I have also worked as an Assistant Project Engineer in a DST, Govt. of India sponsored project titled "Online handwriting recognition for Assamese language" at
IIT Guwahati. I completed my
Ph.D. in Signal Processing & Machine Learning from
IIT Guwahati in Assam, India, under the supervision of
Prof. S. R. Mahadeva Prasanna and
Dr. Prithwijit Guha. Previously, I completed my
M.Tech. in Computer Technology from
NIT Raipur in Chattissgarh, India. I completed my
B.Tech. in Information Technology from
Assam University Silchar in Assam, India. My research interests lie in the areas of speech signal processing, machine/deep learning, audio classification/detection, speaker verification, dialect/language identification and vocal mode classification.
Ph.D. Thesis Details
The title of my Ph.D. thesis is "Analysis of Speech and Music Content for Movie Genre Classification". For my Ph.D. work, I tried to perform movie genre classification using only the audio modality of movie trailers. I studied speech and music signals in general and in the context of movies and TV broadcast signals. I developed methods for the efficient discrimination between speech and music. Subsequently, I automatically extracted speech and music information from movie trailer audio using the proposed algorithms. The speech music information obtained from the trailers was used for the genre classification of movies.