A multi-modal approach to creating routines for smart speakers

Smart speakers can execute user-defined routines, namely, sequences of actions triggered by specific events or conditions. This paper presents a new approach to the creation of routines, which leverages the multi-modal features (vision, speech, and touch) offered by Amazon Alexa running on Echo Show devices. It then illustrates how end users found easier to create routines with the proposed approach than with the usual interaction with the Alexa app.