Tacotron Model and CNN in Virtual Reality for Cancer Diagnosis and Communication between Doctors and Patients

Virtual reality is widely used in various fields, such as military, industrial and medical fields. CNN is prevalent when solving problems about image classification. NLP is the most useful model to realize speech synthesis. However, image classification is seldom combined with speech synthesis to be used in a realistic scene. In order to tackle this issue, this paper proposed a system, which combines image classification with speech synthesis organically. There are three steps used to build this system in this paper. First, this paper devises a model to classify whether the patients have skin cancer. It designs a CNN model to deal with the classification of images of skin, diagnosing whether the patient suffers from Melanoma. Second, the Tacotron model is included in this paper to implement speech synthesis, telling the diagnostic results obtained from image classification to the patients. Finally, a virtual reality environment is built to display a scene when a patient entering the hospital and then being diagnosed and getting treatment.