A speaker recognition based approach for identifying voice spammer

Authors:
Fei Wang;KeXing Yan;Min Feng
Affiliations:
Department of Electronics and Information Engineering, Huazhong University of Science and Technology, Wuhan, China;Department of Electronics and Information Engineering, Huazhong University of Science and Technology, Wuhan, China;Department of Electronics and Information Engineering, Huazhong University of Science and Technology, Wuhan, China
Venue:
WISM'12 Proceedings of the 2012 international conference on Web Information Systems and Mining
Year:
2012

Citing 7
Cited 0

Spoken Language Processing: A Guide to Theory, Algorithm, and System Development

Spoken Language Processing: A Guide to Theory, Algorithm, and System Development
Detecting spam in VoIP networks

SRUTI'05 Proceedings of the Steps to Reducing Unwanted Traffic on the Internet on Steps to Reducing Unwanted Traffic on the Internet Workshop
Adaptive Voice Spam Control with User Behavior Analysis

HPCC '09 Proceedings of the 2009 11th IEEE International Conference on High Performance Computing and Communications
Using Call Patterns to Detect Unwanted Communication Callers

SAINT '09 Proceedings of the 2009 Ninth Annual International Symposium on Applications and the Internet
Collaborative Reputation-based Voice Spam Filtering

DEXA '09 Proceedings of the 2009 20th International Workshop on Database and Expert Systems Application
Efficient Speaker Recognition Using Approximated Cross Entropy (ACE)

IEEE Transactions on Audio, Speech, and Language Processing
Real-time speaker identification and verification

IEEE Transactions on Audio, Speech, and Language Processing

Quantified Score

Hi-index	0.00

Visualization

Abstract

With the blooming of personal communication, more and more people have to process irksome spam calls. However, conventional approaches can hardly gain satisfactory effect when spammers change their phone number or ID randomly. In this paper, we proposed a text-independent speaker recognition algorithm to identify spammer for large population scenario. Firstly, we discover that the codebooks from same person's different speeches are similar. Secondly, we design a vector quantization based CodeBook Filtering algorithm. The experiment results show that our algorithm could achieve a speed-up factor of 30.8:1 with an identification accuracy of 94.98%.