Research Area:  Machine Learning

Abstract:

Humans have always attempted to correspond with objects in a natural language. Communications have been the essential feature of human life, a powerful tool for sharing and building the information that is passed from generation to generation. Among speech processing problems, automatic speech recognition mechanisms of converting the recorded speech signals into the text are one of the most challenging tasks. The signals are typically processed in a digital representation, so speech processing can be observed as a particular case of digital signal processing. The overall performance of an automatic speech recognition system greatly depends upon the acoustic modeling. Hence, building a precise and robust acoustic model holds the key to a suitable recognition performance. People have used different methods for automated speech recognition system. For recognizing the speech people always choose the English language in the majority of the research and implementation but very less work is done in other languages. Our analysis presents the study of the different speech recognition systems present in Indian and foreign languages in the systematic review of speech recognition paper. This paper gives the review of different aspects related to Automatic Speech recognition. We have elaborated the recent advancement in the speech recognition system, robust method for the development of an automatic speech recognition system and application of automatic speech recognition system in different fields.

Keywords:  

Author(s) Name:   Yogesh Kumar; Navdeep Singh

Publisher name:  IEEE

DOI:   10.1109/ICACTM.2019.8776714

Volume Information:  

Paper Link:   https://ieeexplore.ieee.org/abstract/document/8776714