Comment by kiproping

I know google has general sound classifiers like Yamnet, trained on youtube data but they are not very good for specific usecases. So you would have to create a custom model for you usecase.

- https://www.tensorflow.org/hub/tutorials/yamnet