"RUHA" a triggering word used for the model. This model gets the voice from the user and predicts the context/genre in text. Voice dataset is attached. Model is trained by extracting MFCC features of the voices.
• Train any 5 classifiers on the provided dataset using sklearn library. Classifiers can be of your own choice. As you’ve already worked with the dataset and applied KNN on it you must know how to work with audio dataset and how to train a classifier for it. • Display the confusion matrix, accuracy, recall, precision and F1-meausre for the 5 classifiers you have trained. • Now comes the most important part of the project, you will be working with Flask in this project. Flask is a Python web framework that makes it easy to create a fully-featured web application. It is quite simple, easy to learn and is a very demanding skill in the market thus will help you a lot. • You have to create a web application and deploy your project on it, basically interacting on the web application for input and output. A user will be giving the test case on your website e.g., “Ruha lights band kardo” after which using the 5 classifiers you’ve trained you will get a result which you have to creatively display on the webpage. • After the result is displayed, you’ll also have a result analytics button that would redirect the user to another webpage on which the complete result would be displayed. Like, how did the 5 classifiers perform on the dataset.