This Voice Chatbot entertains the user with jokes , facts and an introduction of itself . It has an option between Male and Female voices alongside the generation of text recieved by the Voice Chatbot .This Voice Chatbot was trained on a dataset which contains categories (intents), pattern and responses . A special kind of RNN called LSTM (Longs Short Term Memory) which is capable learning long-term dependencies, especially in sequence prediction problems . LSTM here will classify which category the user’s message belongs to and then we will give a random response from the list of responses. Kivy was used as the framework for the UI and the entire LSTM model along with its train and test was implemented using Keras , Tensorflow , nltk , numpy and pickle.