Introduction

In the design of modern customer experiences, the ability to understand natural language is essential. The Dialogflow component allows you to connect your wolkvox flows with Google's powerful platform (now part of Google Cloud Contact Center AI). This tool processes what the user says or writes, identifies their intent, and returns an intelligent response.

This component acts as a bridge: it sends the user's message to Google and receives the processed response to continue the flow. It is located in the "Cognitive" components tab and is available for all types of routing points (Voice, Chat, etc.).

Configuration

The configuration will depend on the API version you are using in the Google console (V1 or V2).

Configuration for API V1 (Legacy)

If your agent uses the classic version, you will see the following field structure:

• Token: In this field, you must paste the client access token provided by the Dialogflow console.
• Message to send: Here you must enter the variable containing the text the user said or wrote (for example, the variable captured by the ASR or the incoming chat).
• Return Variables:
  • Received message variable: By default, you will see $dialogflow_speech. This variable will store the textual response that the Google bot sends back to the customer.
  • Variable to store intent: By default, you will see $dialogflow_intentName. This variable saves the name of the detected "intent," which is useful for making logical decisions in the flow.
• Once the authentication data and variables are completed, click the "Save Dialog Flow" button at the bottom of the panel to apply the integration.
• After this component, your flow should have an audio playback component (TTS) or a text sending component that uses the $dialogflow_speech variable so the user can hear or read the response generated by Google.

Configuration for API V2 (Standard/Enterprise)

This is the most robust and recommended version. The required fields are:

• Project-id: Enter the unique identifier of your Google Cloud project (Project ID) where the agent is hosted.
• Authentication Field (JSON key file): Below the project ID, you will see a field (masked as a password) where you must place the content of the JSON key file (Service Account Key) generated in Google Cloud.
• Message to send: Enter the variable with the user input you wish to analyze here.
• Return Variables (Predefined):
  • Received message variable: This remains as $dialogflow_speech. It cannot be changed.
  • Variable to store intent: This remains as $dialogflow_intentName. It cannot be changed.
• Once the authentication data and variables are completed, click the "Save Dialog Flow" button at the bottom of the panel to apply the integration.
• After this component, your flow should have an audio playback component (TTS) or a text sending component that uses the $dialogflow_speech variable so the user can hear or read the response generated by Google.