Predicting Human Behavior by Voice and Face Modulation

The secret to clearly expressing one's views and ideas is communication. Speech is the most prevalent and efficient means for humans to communicate out of all the many types of communication. The Internet of Things (IoT) age is evolving quickly, making increasingly sophisticated technology accessible for daily use. Basic wearable gadgets to sophisticated selfdriving cars automated systems used in a variety of industries are examples of these applications. Intelligent apps mostly employ voice- based input, are interactive, and require little user effort to operate. Because of this, these computer programs must be able to fully understand human speech. The speaker's age, race, gender, language, and mood may all be inferred from a speech percept. An emotion detection system is integrated with a number of current voice recognition systems utilised in Internet of Things applications to assess the speaker's emotional state. One of these techniques for identifying human emotions (such as fear, neutral, anger, pleasure, disgust, sorrow, and surprise) is based on the technique for analysing facial expressions employing the popular deep learning