Speech Recognition Interface for Updating Electronic Medical Records with Automatic Itemization

In this study, we propose a voice-based interface designed to enhance the efficiency of populating electronic medical records by healthcare professionals. The interface utilizes advanced voice recognition technology to interpret the verbal input of medical personnel, then leverages a large-scale language model to correctly format the recognized sentences to match for the corresponding sections of the patient's electronic medical record. This approach should expedite the process of updating electronic medical records by eliminating the need for manual data entry. Three variations of our interface, with various strengths and weaknesses are proposed.