As part of the gamification experience of our app, we wanted to incorporate the ability to cast spells. The purpose of these spells were to gain an advantage over what was seemingly a superior opponent. Using spellcasting in a creative way enabled the user to be creative and test various spells to see their impact on the game.

This feature was solved by creating a Power Automate flow. The flow utlized the Azure Cognitive Services API and its Text-to-speech service. The user would then be prompted to use the phone microphone to record a spell, and the flow would then send back a text version and response based on the chosen spell. In addition, the flow also utilized an open API for converting the sound file to the appropriate format for further processing. Below is the outlined flow:

Below is also a demo video, illustrating the feature in effect: