Google Speech to text console Model adaptation configuration for Accuracy Enhancement

I am trying to use Google Speech to Text transcription service on its console. There is a Model adaptation field to enhance the accuracy. However, even I have created several phrase sets in my configuration, I still can not see any of them in the Model adaptation as saved adaptation resource for choosing. I don't know what I missed, I need some one expert in Google Speech to Text service to show me how to configure the Model adaption. The requirement of this project is very specific and hands-on experience only, if you don't know Google Speech to Text service console or never used it then please don't apply. Thanks!