Speech emotion recognition is a technique used to identify and understand the emotions conveyed in spoken language. Deep learning, a subset of machine learning, has been proven to be an effective method for speech-emotion recognition. This is due to its ability to learn and improve upon a wide range of features in speech, such as MFCC, pitch, intonation, and rhythm. In this abstract, we present a deep learning approach to speech emotion recognition, which utilizes a combination of recurrent neural networks (RNNs) and convolutional neural networks (CNNs) to accurately classify emotions in speech. The Speech Corpora dataset is used in speech emotion Identification and gets a Taining Accuracy of 86.25% For the Recurrent Neural Networks (RNN) Model. This paper aims to explore the use of deep learning techniques for speech emotion recognition. We begin by reviewing the current state of the field, highlighting recent advances and key challenges. We then describe the deep learning models that have been most commonly used for speech emotion recognition, such as convolutional neural networks (CNNs) and recurrent neural networks (RNNs). We discuss the advantages and limitations of these models and the various pre-processing and feature extraction methods that have been proposed. Finally, we present recent results and evaluate the performance of deep learning-based speech emotion recognition methods, and conclude with some future directions for the field.

Emotion Recognition, Deep Learning, Recurrent Neural Network (RNN), LSTM.