Swarm Intelligence based computing techniques in speech enhancement
- Resource Type
- Conference
- Authors
- Devi, Khumukcham Usharani; Sarma, Dipjyoti; Laishram, Romesh
- Source
- 2015 International Conference on Green Computing and Internet of Things (ICGCIoT) Green Computing and Internet of Things (ICGCIoT), 2015 International Conference on. :1199-1203 Oct, 2015
- Subject
- Communication, Networking and Broadcast Technologies
Components, Circuits, Devices and Systems
Computing and Processing
Power, Energy and Industry Applications
Robotics and Control Systems
Signal Processing and Analysis
Transportation
Speech
Algorithm design and analysis
Adaptive filters
Speech enhancement
Signal to noise ratio
Finite impulse response filters
Filtering algorithms
LMS
RLS
PSO
ABC
SNR
- Language
Speech enhancement has become an important topic in Digital Signal Processing Systems, as the main problem in speech enhancement is the presence of background noise which leads to the degradation of the quality of the speech signals. Removal of background noise and echo suppression has become necessary so that the intelligibility of the speech signal is improved. This may be achieved by adaptive filter. In this work, a comparative analysis between the Gradient based algorithms that is the LMS (least mean square) and RLS (recursive least square) and the swarm intelligence based global optimization algorithm such as PSO (particle swarm optimization) and ABC (Artificial Bee Colony) is discussed. The experimental results shows that swarm Intelligence based optimization algorithm techniques appears to give a better SNR(Signal to Noise Ratio) improvement than the conventional gradient based algorithm.