how to create the training data?

hallo.

I have folder TrainData. in this folder I have 5 folder 'cat' , 'dog', 'fish', 'lion', 'tiger'. In each folder i have 81 sample audio.

I will make the speech recognition with mfcc feature extraction. But I don't understand about training data for first step.

Anyone can help me, how to create the program for data training?

thanks

[Walter Roberson]

all of that data is training data. If it were test data or validation data, then the folder should have been named something different.

[nana rosd]

ooh okay thank you.

Answer 1

[jibrahim]

Nana,

Take a look at the 'Speaker Identification Using Pitch and MFCC' example in Audio Toolbox. You can probably do something similar for your problem.

https://www.mathworks.com/help/audio/examples/speaker-identification-using-pitch-and-mfcc.html

[nana rosd]

yes, of course. I will try it. thanks