How to create your own training base?

[Pidberezkiy]

Hello, I just started to intercess in this, and a week later I managed to launch the project on the Android studio and get applications for installations. But I can’t figure out how to create a JSON classification for audio https://github.com/araobp/android-aed/tree/master/data if I want to use other sounds for classification, such as crying a child, or something else. Can you explain to me how to do it? I assume that the answer is here https://haythamfayek.com/2016/04/21/speech-processing-for-machine-learning.html but for me it’s difficult for me. Thank you.

[Pidberezkiy]

You should know that you can film this on your phone. «/Android/data/audio.processing.spectrogram/files/Music». But I would still like the same thing on a computer. And I also have problems with Jupyter notebook. After passing through epochs at some point, I get an error, and after that the model is saved with a size of 0 bytes. PS. It’s great, I managed to solve the problem with notebook, I launched it through Google Colab. The only question was how to prepare a dataset from audio on a computer, and create tags in these .JSON