Skip to main content
Back to External NLU Integrations The Dialogflow ES adapter uses NLP and machine learning to identify user intent and extract entities. This page covers how to create and train a Dialogflow agent, enable the Dialogflow API, and generate service account credentials for use in the Kore.ai External NLU adapter settings. Once configured, you can test the conversation behavior of your AI Agent using your Dialogflow ES model.

Train the Dialogflow Engine

Prerequisites: At least one agent created on Dialogflow Essentials. Intent and entity names in Dialogflow must match those defined in the Kore.ai XO Platform. To configure the Dialogflow ES adapter, you need the Project ID and Private Key from your Google Cloud service account JSON file. See Enable the Dialogflow API and Create Service Account Credentials.

Create and Train an Agent

  1. Go to Dialogflow Essentials and click Create Agent. In this example, the agent is named Test_Dialogflow. Create an agent
  2. Click Create Intent and name it to match the intent in your AI Agent. Create an intent with the same name as in your App
  3. In this example, the intent is named Phone Number Verification to match the corresponding intent in the AI Agent.
  4. Under Action and Parameters, add the action name matching the entity, then fill in Parameter Name, Entity, Value, and Prompt. Add action, parameter, entity, value, prompt
  5. Under Responses > Default, add the Text Response to display to the user. Add text response
  6. Click Save.
  7. Go to Training in the left menu and enter the intent name under User Says. Go to Training and enter Intent name
  8. In the Try Now panel (top right), enter the intent name to test the configuration.
  9. The configured text response appears under Default Response. Configured text response displayed under Default Response
    You can configure any intents, entities, or FAQ details you need to identify using the Dialogflow ES adapter.

Enable the Dialogflow API

  1. Click the Gear icon next to the agent name in the Dialogflow dashboard. Click the Gear icon next to the agent name
  2. On the General Settings page, scroll down to find the Project ID. Note the project ID
    The Project ID is required when configuring the Dialogflow ES adapter on the Kore.ai platform.
  3. Click the Project ID to open Google Cloud.
  4. In the Google Cloud dashboard, go to APIs & Services > Enabled APIs & Services.
  5. Click + Enable APIs and Services and search for Dialogflow API. Click Enable APIs and Services Find Dialogflow API option
  6. After activating the Dialogflow API, click Manage to open the API management dashboard. Dialogflow API management dashboard

Create Service Account Credentials

Create a Service Account

  1. Click Create Credentials at the top of the Dialogflow API management dashboard.
    If you already have service accounts, access them from Credentials in the left navigation menu.
    Click 'Create Credentials'
  2. On the Create Credentials page, Dialogflow API is pre-selected in the Select an API dropdown. Dialogflow API is by default selected
  3. For What data will you be accessing?, select Application Data. Click Next. Application Data selected
  4. Select No, I’m not using them and click Next. No, I'm not using them
  5. Fill in the service account name, ID, and description. Service account details
  6. Click Create and Continue.
  7. Grant the required access permissions and user access on the subsequent pages. Grant permission to Service Account Grant permission to User

Generate a Private Key

  1. Click the service account under Service Accounts in APIs & Services. Click on the Service account
  2. Go to the Keys tab. Service account details - Keys tab
  3. In the Add Key dropdown, select Create a new key. Create a new private key
  4. Select JSON and confirm. JSON option to create Private key
    JSON is selected by default.
A JSON file is downloaded to your computer. JSON Private Key information downloaded The file structure:
{
  "type": "***",
  "project_id": "***",
  "private_key_id": "***",
  "private_key": "***",
  "client_email": "***",
  "client_id": "***",
  "auth_uri": "***",
  "token_uri": "***",
  "auth_provider_x509_cert_url": "***",
  "client_x509_cert_url": "***"
}
Save this JSON file securely — you cannot access it again. Copy the entire contents into the Private Key field in the Kore.ai External NLU adapter settings.

To validate the Dialogflow NLU configuration, see Test and Debug.