International Journal of Computer Applications |
Foundation of Computer Science (FCS), NY, USA |
Volume 63 - Number 9 |
Year of Publication: 2013 |
Authors: Deeksha Bhatnagar, Vikash Singh, Sandip Vijay |
10.5120/10496-5253 |
Deeksha Bhatnagar, Vikash Singh, Sandip Vijay . Database Enhancement and Analysis of Spoken Hindi Reduplicated Words using Endpoint Detection Algorithm. International Journal of Computer Applications. 63, 9 ( February 2013), 30-35. DOI=10.5120/10496-5253
Speech recognition is basically pattern classification task. It is divided mainly into two major parts. The first part is speech signal processing and the second part is speech pattern recognition technique. This paper describes the database expansion & improvement of spoken Hindi Reduplicated words. The main focus is on the analysis of Spoken Hindi Reduplicated words by using endpoint detection algorithm. The improvement is done in the robust environment. Male and female speakers are selected from different age- group to explore the acoustical variation in their utterances. Various emotions such as normal, happy, anger and surprise are taken for analysis of speech signal. Endpoint detection algorithm is used for the analysis purpose of various parameters such as total duration of time, number of samples, root mean square value and mean power (intensity) in air of speech signal.