How to train a neural network having a dataset with 5 sound files?

Question

Python

Hello everyone, I am making a model based on https://www.tensorflow.org/tutorials/sequences/aud... according to the tutorial, we are given a 2 GB dataset, but I have only 5 voice recordings of one word, for example cat, Is this tutorial suitable for me, if so, how to pull up train.py under a small dataset.
I only need to recognize the word cat, like Hello Siri or Ok Google
I can’t use someone else’s api, I need to get tfile on the output, then most likely convert to tlfile

Answer 1

Why do you need a neural network for this? There is a SpeechRecognition library that does an excellent job of recognizing voice.
What kind of fashion has gone to shove neurons everywhere. Did you by any chance try to attach a neuron to the parser in order to search for divs and tables?