Personal Assistant with Emotion Recognition Based on Artificial Intelligence
N. Kumaran1, Kavyalakshmi E V2, Maithreyi S3
1N Kumaran*, Student, Department of Computer Science and Engineering, SCSVMV University, Kanchipuram. (Tamil Nadu), India.
2KavyaLakshmi E V, Student, Department of Computer Science and Engineering, SCSVMV University, Kanchipuram. (Tamil Nadu), India.
3Maithreyi S, Student, Department of Computer Science and Engineering, SCSVMV University, Kanchipuram. (Tamil Nadu), India.
Manuscript received on March 03, 2022. | Revised Manuscript received on March 08, 2022. | Manuscript published on March 30, 2022. | PP: 67-70 | Retrieval Number: 100.1/ijitee.C98120311422 | DOI: 10.35940/ijitee.C9812.0311422
Open Access | Ethics and Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: This paper deals with a method adapted to create a personal assistant based on emotion recognition. It explains how the emotions of users can be recognized and suggests the relevant methods. Now, we have an abundant number of personal assistants and chatbots: we took one of such bots a step ahead and made our assistant to automatically recognize users’ emotions. Through this, our assistant is made to acquire empathy and usage with a lot of fun and advantage. The application has the potential to be used in mobile phones. Many people suffer from social anxiety and depression; our assistant is intended to help them by identifying their emotions and by suggesting ways to cheer them up. The assistant induces the user to be more connected to their phone and also to other people. Initially, our idea was to create a futuristic application, but later this turned out to be more than that. It proves to be advantageous for all mobile phone users. We created a simple personal assistant (Edith) with emotion-sensing ability. We used deep learning techniques to detect users’ emotions. We illustrated our model with a lot of human faces displaying different kinds of emotions. Achieving a decent accuracy while recognizing the emotion was the hardest part in this attempt because very slight differences in facial muscles portray dissimilar emotions; identifying the pattern in similar feelings is challenging. We overcame these difficulties by choosing an appropriate dataset with many images to depict our model to recognize the emotions correctly. Our model achieved better accuracy by doing so.
Keywords: Edith, Emotion Recognition, Personal Assistant, Virtual Assistant.
Scope of the Article: Artificial Intelligence