Linear Prediction Analysis and Spectrographic Depiction

Generally, in recent research, two methods are used to estimate the formant form of vowel. Linear prediction analysis (LP analysis, or synonymous linear prediction coding [LPC]) and spectral description. In addition, many studies have combined these two methods by cross-checking these values ​​by numerically computing the frequency, bandwidth, and formant amplitude through LP analysis and visually examining the relevant spectrograms It was. LP analysis relies on source filter theory of speech generation.

Linear regression analysis is the most widely used of all statistical methods. It is a study of linear additive relationships between variables. Suppose Y represents the "dependent" variable whose value you want to predict, and X1, ..., Xk represents the "independent" variable you want to predict. Here, the value of the variable Xi is within the period t (or row t) by Xit. The equation has the following characteristics: Y prediction is a linear function of each X variable, other variables remain fixed, and contributions of different X variables to the prediction are as follows. It is an additive. The slope of the linear relationship with Y is the constant b 1, b 2, ..., b k, the so-called variable coefficient. In other words, bi is the change in predicted value of Y per unit change in Xi, the other conditions are the same.

In statistics, linear regression is a method of predicting target variables by fitting the optimal linear relationship between dependent variables and independent variables. Optimizing fitting is done by ensuring that the sum of all the distances between the shape at each point and the actual observed value is as small as possible. In the case of shape selection, the error is reduced at other positions, so the fit of the shape is "best". The two main linear regressions are simple linear regression and multiple linear regression. Simple linear regression predicts the dependent variable by approximating the optimal linear relationship using a single independent variable. Multiple linear regression predicts a dependent variable by using multiple independent variables and approximating the best linear relationship.

Linear regression is a general statistical analysis technique. It is used to determine the degree to which a dependent variable has a linear relationship with one or more independent variables. There are two types of linear regression, simple linear regression and multiple linear regression. Simple linear regression uses a single independent variable to predict the value of a dependent variable. In multiple regression two or more independent variables are used to predict the value of the dependent variable. The difference between them is the number of independent variables. In either case, there is only one dependent variable