Description

In this lab assignment, we have extracted the audio features from the given input audio samples and these extracted features are later used for Audio Classification. In the below screenshot, we took few audio samples like bell, claps and fire alarm sound as the input and extracted their features into a text file i.e. Features.txt.

Feature Extraction Output Screenshot:

Here, we use Spark machine learning libraries for processing the extracted features. The extracted features are taken as the training dataset. We used decision tree algorithm to build our model. We validated our model using the test dataset. Finally, our model successfully predicted the input data set.

Audio Classification Screenshot:

Later, the Classification Model will be deployed into Mongo Lab and further audio processing will be done using Storm.