Vox Ludorum is a speech-to-text-to-speech module for FoundryVTT powered by Azure Speech
Check out a short demo video here: https://user-images.githubusercontent.com/15639841/114201277-a0732700-991b-11eb-98e5-a05d3365e944.mp4
Due to how modern browsers protect you, Media sources (both microphone and speaker) are unavailable when on insecure connections such as HTTP. This module will not work on direct IP connections such as http://[IP]:30000/game
Vox Ludorum could definitely do more than it does now, but I wanted to get out what I had in a usable format rather than let it never be released at all. Unfortunately, this is being released as-is: PR's are welcome, but I do not anticipate having any time to add any new features to this.
https://azure.microsoft.com/en-us/free/cognitive-services/
Speech to Text - 5 audio hours free per month, $1 / hour after that Neural Text to Speech - 0.5 million characters free per month, $16 per 1M characters after that