I'm trying to train and use a neural network to detect a specific word in an audio file. The input of the neural network is an audio of 2-3 seconds duration, and the neural network must determine whether the input audio (the voice of a person) contains the word "hello" or not.

I do not know what kind of network to use. I used the SOM network, but I did not get the desired result. My training data contains a large number of voices that contain the word "hello".

Is there any python code for dis problem?

  • $\begingroup$ You can use a CNN or LSTM model. For the CNN one you can try pre-trained models for starting and for the LSTM one you can try bidirectional LSTMs. $\endgroup$
    – Let's try
    Aug 3, 2020 at 9:42
  • $\begingroup$ What datasheet should I use to train the class related to "not hello"? $\endgroup$ Aug 3, 2020 at 13:08
  • $\begingroup$ Have you tried looking at some models at tensorflow hub? I’m positive there should be some useful stuff $\endgroup$
    – pedrum
    Aug 3, 2020 at 20:02

1 Answer 1


After some research on the internet, I realized that using VOSK toolkit in python, it can be found (detect) any particular word in audio file or real time audio streaming.



You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .