A Review on Forensic Speaker Recognition System |
Author(s): |
| Pardeep Sangwan , Maharaja Surajmal Institute of Technology |
Keywords: |
| Variability, Database, I-Vector, Neural Network, DNN, Deep learning |
Abstract |
|
The voice features like speaking style pitch, fundamental frequency of each speaker is unique. Speech recognition is the capability to recognize said words whereas speaker recognition recognizes the speaker on the specific characteristics obtained from speaker voice. Speaker recognition is a significant area in the field of speech processing. The standard databases existing for speaker recognition and forensic speaker recognition also discussed .The criterion for selection of these database also covered in this study. In speaker recognition various sources of variability has been discussed. Various factor responsible for variability has been explained. This paper describes the I-Vector and deep learning techniques. |
Other Details |
|
Paper ID: IJSRDV7I100345 Published in: Volume : 7, Issue : 10 Publication Date: 01/01/2020 Page(s): 474-479 |
Article Preview |
|
|
|
|
