Speech Recognition using Deep Neural Network

Speech recognition has been an important sector of research to enhance the user interaction with machines and electronic devices. Speech recognition is used commonly in this era, with handheld devices, home automation systems, and many more. There are different methods to creating speech recognition devices, these methods include Hidden Markov Model (HMM) and Gaussian Mixture Model (GMM). However, with recent developments in research it is found that the most accurate method of speech recognition is by implementing the use of Deep Neural Networks (DNN); as it has shown high accuracy in comparison to HMM or GMM models. Deep Neural Networks are commonly used for image recognition, artificial intelligence and more. In this paper, we will discuss the methods that are used for speech recognition, and implement a system using DNNs and understand the parameters that make the difference in the model.