Towards Security Hardening of Scientific Demand-Driven and Pipelined Distributed Computing Systems
ISPDC '08 Proceedings of the 2008 International Symposium on Parallel and Distributed Computing
Choosing best algorithm combinations for speech processing tasks in machine learning using MARF
Canadian AI'08 Proceedings of the Canadian Society for computational studies of intelligence, 21st conference on Advances in artificial intelligence
Simple dynamic key management in SQL randomization
NTMS'09 Proceedings of the 3rd international conference on New technologies, mobility and security
Autonomic specification of self-protection for distributed MARF with ASSL
C3S2E '09 Proceedings of the 2nd Canadian Conference on Computer Science and Software Engineering
Simple dynamic key management in SQL randomization
NTMS'09 Proceedings of the 3rd international conference on New technologies, mobility and security
Evolution of MARF and its NLP framework
Proceedings of the Third C* Conference on Computer Science and Software Engineering
Unifying and refactoring DMF to support concurrent Jini and JMS DMS in GIPSY
Proceedings of the Fifth International C* Conference on Computer Science and Software Engineering
Proceedings of the 2013 Research in Adaptive and Convergent Systems
| Hi-index | 0.00 |
This work reports experimental results and their analysis in various speech processing tasks using SpeakerIdentApp, a text-independent speaker identification application, based on Modular Audio Recognition Framework (MARF)'s API and its implementation in terms of best of the available algorithm configurations for each particular task using median clusters as opposed to the default mean clusters. This study focuses on the tasks of identification of speakers' as of who they are, their gender, and accent through machine learning. This work significantly complements two preceding statistical studies undertaken using only mean clusters and shows the difference in selection of the best algorithm combinations using the median cluster approach. To the author's knowledge there was no any previous comprehensive study in this regard.