A Comparative Study on Feature Extraction Technique for Isolated Word Speech Recognition
Journal: International Journal of Engineering and Techniques (Vol.1, No. 6)Publication Date: 2015-11-01
Authors : Easwari.N; Ponmuthuramalingam.P;
Page : 104-107
Keywords : Automatic Speech Recognition; MFCC; RASTA; Isolated Word; Speech Chain; Signal Noise Ratio.;
Abstract
One of the common and easier techniques of feature extraction is Mel Frequency Cestrum Coefficient (MFCC) which allows the signals to extract the feature vector. It is used by Dynamic Feature Extraction and provide high performance rate when compared to previous technique like LPC. But one of the major drawbacks in this technique is robustness. Another feature extraction technique is Relative Spectral (RASTA). In effect the RASTA filter band passes each feature coefficient and in both the log spectral and the Spectral domains appear linear channel distortions as an additive constant. The high-pass portions of the equivalent band pass filter effect the convolution noise introduced in the channel. The low-pass filtering helps in smoothing frame to frame spectral changes. Compared to MFCC feature extraction technique, RASTA filtering reduces the impact of the noise in signals and provides high robustness
Other Latest Articles
- A Review of Transmission System for Performance of CVT
- A Review on Mobile AD HOC Networks
- Analysis of Leakage in Bolted-Flanged Joints with Gasket and Under Thermal Condition, a Critical Review
- A Capable Text Data Mining Using in Artificial Neural Network
- A Review on Image InpaintingTechniques and Its analysis
Last modified: 2016-01-10 11:20:07