Cyber-Ego: The Digital Self

Cyber Ego expands to Cyber and Alter-ego. In this proposed paper, I have integrated four deep learning models for performing the tasks - natural language generation, text-to-speech, speech-driven facial animation and image animation as a combined framework to generates a video of the character speaking the generated content with the character’s voice. Text data and audio voice lines have been collected and pre-processed to train the text-to-speech model. While pre-trained models are fine-tuned and used for the speech-driven facial animation model and image animation model. To put everything together in an interactive way, a Flask-based Python web application is created to provide as an interface that handles user inputs and generates the result.