People are known to have different personality types, each corresponding to its own typical behaviors and preferences. Personality type can influence how an individual perceives received messages, e.g., voice-based responses from a digital assistant. However, current digital assistants provide speech out that is not customized to the user’s personality. This disclosure presents techniques for applications that provide voice-based user interact to adapt the messages communicated to a user to suit the personal characteristics of the user. User’s personality is inferred with specific user permission and is based on one or more user-permitted factors. For users that do not provide permission or where the factors are insufficient, default messages are used. A trained machine learning model is utilized to infer user personality and preferred communication style based on the user-permitted factors. The inferred style is then applied to adapt information content and delivery via a voice user interface.

Creative Commons License

Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.