In software engineering, the adoption of mobile devices generates the need for proposing new methods and processes for developing mobile applications. This paper presents and describes a software development process for the automatic generation of medical mobile applications using voice recognition. The software development process is supported by a conversational agent that captures the functional requirements of the application to be developed. A Web-based prototype was developed as a proof-of-concept of the process proposed. We believe this process will allow reducing time and effort to developers in mobile applications.