Classify Sound Using Deep Learning database how ?

I tried to do my Classify Sound Using Deep Learning with some words and I wan to do this

with this example

https://www.mathworks.com/help/audio/gs/classify-sound-using-deep-learning.html?searchHighlight=speech%20recognition&s_tid=doc_srchtitle

but i dont know how to create my own database and where to save it ? and what i must change in code. Please help.

Answer 1

[Krishna]

Hello,

Seems like you want to build an audio recognition system, for that you need to learn where to store the dataset and how to access it.

For such deep learning task where we need to have large dataset and we cannot load dataset into the workspace, MATLAB has defined datastore which can access the data directly from the location without needing to bring it into workspace.

For audio we have specific ‘audioDatastore’ object to create and manage audio datasets. Please look at the following documentation,

https://in.mathworks.com/help/audio/ref/audiodatastore.html

To learn how to manage and retrieve data using these datastores and to build audio recognition systems using deep learning look at the following examples,

https://in.mathworks.com/help/audio/ug/load-and-manage-audio-data-sets.html

Also to know how to extract features from audio data sets look at the following example,

https://in.mathworks.com/help/audio/ug/extract-features-from-audio-data-sets.html

Hope this helps.