This article explains how you can connect the babelforce AI Agent with one of your Zendesk AI Agents (previously Ultimate bots).
You can only use the babelforce AI Agent if you get in touch with the babelforce Success Team. They will add your API key to your account. Also, you will need to authenticate your bot against the babelforce platform. We will explain this a bit later.
Before connecting your Zendesk AI Agent flow with the babelforce AI Agent, you need to follow some setup steps in your Zendesk account. Also, you have to get in touch with the Success team, they will share two items with you:
- Webhook URL
- A Bearer Token
And these are the steps you need to take to connect your Bot with the babelforce AI Agent:
- Find your AI agents' Dashboard and click on Custom CRMs.
- Provide details about the integration, like Name and Organization
-
Now enter the Webhook URL (1. in the screenshot)
- It will be something like this, depending on your environment:
https://myEnv.
babelforce.com/ivr/api/v1/webhooks/ultimate/chat
- It will be something like this, depending on your environment:
-
Finally, add one header (2. in the screenshot):
- Key = Authorization
- Value = Bearer + Token whereas the token was provided to you by the babelforce Success team
- Save your new Custom CRM.
Click "Select your CRM" and chose the custom CRM you created from the drop down list.
Make sure to regenerate the Access token. Please share this token with the Success Team as it needs to be saved in your account's vault in order to enable the bot communication.
The Access token will only be displayed once.
Finally, make sure to create a welcome message in your bot, otherwise the customer calling your hotline will not be welcomed by a prompt.
In your AI agents' Dashboard go to Content > Use cases and add the Welcome reply for your service.
Once you configured the steps in Zendesk, ensure the Success team receives both the api-key as well as the bot id.
Wrapping up in babelforce
To complete the setup, create a new AI Module via the babelforce Manager app.
When you setup the agent, select the input language and voice of the bot.
Most importantly, add the botId as Request body like this:
botId
followed by the id in the next field.
Set a waiting audio and decide on a failure audio. Latter is a fallback in case the bot takes too long or does not have an answer to the question.
As exit condition, you can decide how long a customer is allowed to stay within the bot as a maximum. You can also decide how many iterations the customer can have with the bot.
Remember to define a flow that follows after the bot conversation is finished.
Related to
Comments
0 comments
Please sign in to leave a comment.