Frequency Tremor Intensity Index (FTRI)

10 次查看(过去 30 天)
Hello, currently I have a task to extract MDVP (Multi-dimentional Voice Parameter) from sample voice.
I really new using matlab for doing voice processing. I found some problem to extract FTRI, I was given the step to extract it, but I don't know how to interpret the step into matlab code.
The method for FTRI consist of the following steps:
Division of fundamental frequency period-to-period (Fo) data into 2 sec. windows. For every window, the following procedure apply:
  1. Low-pass filtering of the Fo data at 30 Hz and downsampling to 400 Hz
  2. Calculation of the total energy of the resulting signal.
  3. Subraction of the DC-component.
  4. Calculation of autocorrelation funtion on the residue signal
  5. Division by the total energy and conversion to percent.
  6. Extraction of the period of variation.
  7. Calculation of FFTR and FTRI corresponding to the priod of variation found
Computation of the average autocorrelation curve and average FTRI for all processed windows.
[y,fs] = audioread('sample.wav'); %load audio file
%% Average Fundamental Frequency (FO)
[f0, idx] = pitch(p, fs);
FO = mean(f0(:,1));
I only know the way only to get Fo value.
If anyone know more about this one, please help me, and I anyone know about voice processing tools provided by matlab that can extract easily those MDVP feature please tell me.
Thank you,
Best Regards.
  1 个评论
Epri Pratiwi
Epri Pratiwi 2020-4-22
I have the similar case before, perhaps you can try this code. Good luck!
clear; close all; clc;
FS = 16e3;
[x,fs] = audioread('S_01_01.wav');
x = resample(x, FS, fs);
y=vocoder(x, FS, 8, 160, 'TONE', 1);
figure;
plot((1:length(y))/FS, y);
xlabel('Time (second)')
title('Vocoded signal')

请先登录,再进行评论。

回答(0 个)

类别

Help CenterFile Exchange 中查找有关 Spectral Measurements 的更多信息

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by