AUTOMATIC INITIAL SEGMENTATION OF SPEECH SIGNAL BASED ON SYMMETRIC MATRIX OF DISTANCES
Journal: INTERNATIONAL JOURNAL OF COMPUTERS & TECHNOLOGY (Vol.13, No. 9)Publication Date: 2014-08-09
Authors : Dmytro Peleshko; Yuriy Pelekh; Mariya Rashkevych; Yuriy Ivanov; Iryna Verbenko;
Page : 4782-4791
Keywords : Speech signal; Initial segmentation; Symmetric matrix of distances; Pseudo inverse matrix; Moore-Penrose algorithm; Singular decomposition; Euclidean; Chebyshev and cosine metrics; Measure of similarity.;
Abstract
The most common issue of a speech signals analisys and artificial intelligence systems development is determining of temporal and frequency charactristics. That’s because any undetermined signal is defined as a nonlinear object. But it is always possible to select time interval from the signals with a given discretization period. Such an interval is called quasistationary interval.The quasistationary interval in combination with the speech signal quality characteristics can be used to build a parametric model of the speech signal. As a result, such a model will be very helpful for solving different issues of artificial intelligence development processes.In this paper the method of an initial segmentation of the speech signal via quasistationary intervals by threshold deviation of algebraic specifications of elementary areas is presented. The latter method of automatic initial segmentation is based on pseudorotation of symmetric matrix of distances and evaluation criteria. The problem and process of selection the quasistationary intervals are described in the section 1.
Other Latest Articles
- A Hybrid Multi-Word Terms Extraction System Applied to Topic Detection
- Content-Based Image Retrieval using Color Quantization and Angle Representation
- Investigating the Synergistic Relationship between Enterprise Resource Planning and Business Intelligence
- Highly Scalable Network Management Solution Using Cassandra
- A Survey on Intelligent Water Drop Algorithm
Last modified: 2016-06-29 16:54:52