Service Keys

Parloa's integration with Dialogflow-based platforms, including Google Assistant, Phone, and WhatsApp, offers a seamless automation feature for uploading and updating speech assets. This streamlined process enhances efficiency by ensuring that speech assets are consistently synchronized across your conversational AI ecosystem, from development to debugging.

Setting Up Automatic Uploads for Speech Assets

The auto-upload feature in Parloa works for both releases and debugger instances. Here's how to set it up for your Dialogflow agents:

  1. Access your list of Dialogflow agents and select the agent you wish to configure, or create a new one.

  1. Click on the associated Service Account for your chosen agent.

You'll be redirected to the Google Cloud Platform.

  1. If necessary, set up a role with specific permissions by navigating to the Roles section:

  1. Create a new role by clicking on + CREATE ROLE. Assign a name and fill in all required fields.

  1. Add necessary permissions by clicking + ADD PERMISSIONS.

In the modal that appears, filter for the "Permission" category and input the required permissions. These include dialogflow.agents.get, dialogflow.agents.create, and several others related to Dialogflow and, if needed, TENIOS for voice bot setup.

These are the required permissions:

  • dialogflow.agents.get

  • dialogflow.agents.create

  • dialogflow.agents.restore

  • dialogflow.agents.train

  • dialogflow.agents.update

If you plan to set up a VoiceBot via TENIOS, the following additional permissions are required:

  • actions.agent.get

  • dialogflow.intents.list

  • dialogflow.sessions.detectIntent

  • dialogflow.sessions.streamingDetectIntent

  1. Select the permissions by typing each one into the filter field, checking the corresponding box, and clicking ADD.

If you follow each permission with an OR (it will be offered once you've selected the first permission), you can even add all of them at once.

  1. After all permissions are added, finalize the role creation by clicking CREATE.

  1. Return to the Service Accounts section to create a new service account, selecting the role you've just created, and continue through the prompts until completion.

Select a name for your service account and click on CREATE:

In the next screen, select the previously created role and click on CONTINUE:

On the next screen simply click on DONE.

Finally, you can create the service key. Simply click on the newly created service account to edit it:

On the next screen, click on ADD KEY and then on Create new key:

Select JSON and confirm by clicking once more on CREATE:

This will trigger the download of the service key. Click on CLOSE and then on DONE:

Now you can go back to your Parloa tab and upload the service key to your environment:

In your releases and debugger instances, simply select your environment to use the auto-upload feature:

Click once on Upload Speech Assets to trigger the initial upload in your release/debugger. But afterward, you will never have to do it again, since any changes to your speech assets will automatically be uploaded from Parloa to your Dialogflow agent.

In the debugger, you can click on the "Eye" icon to copy the fulfilment link. Then, in Dialogflow, you can use the link to communicate with your debugger.

Last updated