Problem in Child Gender Classification using Speech

1 次查看（过去 30 天）
Kodali Radha
Kodali Radha 2021-8-15
评论： Kodali Radha 2021-8-15
Dear all, pls help me out, I'm coding Gender Classification from children speech.
When I extracted MFCC features I have got that in 3D, MFCC=5023x14x2
and pitch=5023x2. Now I want to concatenate both pitch and MFCC for labeling but I'm getting an error here as "dimension of array is not consistent".
So I thought of using this
melC_1=melC(:,1:14); %% Changed dimension of MFCC from 3D to 2D = 5023 x 14
f0_1=f0(:,1); %% Pitch is 5023 x 1
feat = [melC1, f0_1]; %% Concatenated
My actual problem: Am I correct in reducing the dimension like shown above? Because I'm getting very much less accuracy when I use the above concatenation.

请先登录，再回答此问题。

采纳的回答

Image Analyst
Image Analyst 2021-8-15
What does
size(melC_1)
size(f0_1)
show? Don't use a semicolon so that the results will show up in the command window. Evidently the sizes are not what you thought.
If you went from 3-D to 2-D you might need to use squeeze():
melC_1 = squeeze(melC(:, 1 : 14)); %% Changed dimension of MFCC from 3D to 2D = 5023 x 14
f0_1 = f0(:, 1); %% Pitch is 5023 x 1
feat = [melC1, f0_1]; %% Concatenated

更多回答（0 个）

类别

在 Help Center 和 File Exchange 中查找有关 Signal Processing Toolbox 的更多信息

标签

产品


版本

R2021a

Community Treasure Hunt

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

Start Hunting!

Translated by