While reading the action.py I noticed that there is a new action “ActionRetrieveResonse” which can be triggered with the new prefix “respond_”. However, I haven’t yet understood the use of this action and the difference to “utter_” action. Can anyone help?
Hi @matthiask. This action corresponds to the new response selector which we added to Rasa very recently. Daksh, the main person behind this feature also wrote a post on it which you can read to learn all the details about how it works and how it is different from simple utterances: Integrate response retrieval models in assistants built with Rasa