I am curious if anyone has used Misty to detect images with the Google Vision API, and if so how exactly was you able to establish connection? I am struggling to figure that out.
I have not used it, but the following example with Google Cloud Text-to-Speech API has a pattern that you can follow:
I did tried to use Google Vision earlier for the same skill that you are trying to build.
But unfortunately I did not had any luck with that.
One thing to note, Google vision does not support base 64 image, you will need to convert the base 64.