- A Copilot: Next Action Tile node that suggests a next action to the human agent.
- A Copilot: Transcript Tile node that displays the conversation transcript to the human agent.
Prerequisites
- Access to Voice.
- Configure a speech provider in Cognigy.
- Configure the phone number you use in your contact center.
- Configure a contact center platform that you can integrate with a handover provider. For example, for Live Agent, you need to have an Inbox with the Project ID and the Webhook URL set up and, if necessary, the Enable Agent Copilot Workspace option activated.
- Add a compatible LLM for the LLM Prompt node.
Setup process
- Create a handover provider. The handover provider connects the agent to your contact center platform.
- Create a handover flow. The handover flow configures the interaction with the agent and the handover to a human agent.
- Create an Agent Copilot flow. The Agent Copilot flow configures the Agent Copilot workspace and the individual widgets.
- Create a Voice Copilot Endpoint. The Voice Copilot Endpoint receives the transcript of the voice call and sends it to the Agent Copilot workspace.
- Create a Voice endpoint. The Voice endpoint connects the AI Agent to Voice and your contact center phone number.
- Test the Agent Copilot workspace by calling the phone number you assigned to the Voice endpoint.
Set up Agent Copilot for voice-to-chat
1
Create a handover provider
In Deploy > Handover Providers, create a handover provider according to your contact center platform. Depending on the platform, you may need different settings for the handover flow.
2
Create a handover flow
- In Design > Flows, create a flow and give it a clear name, for example,
Handover. - In the flow editor, add a Say node and set its Text parameter to
Hi, let's hand you over to the humans.. - Add a Handover to Human Agent node and configure the following:
- Handover provider — set to the previously created handover provider, and configure it accordingly. For example, if you use Live Agent, you need to set Inbox ID in the Live Agent settings section.
- Handover acceptance message — enter
Transferring to humans....
3
Create an Agent Copilot flow
-
In Design > Flows, create a flow and give it a clear name, for example,
Agent Copilot. -
Add a Copilot: Set Grid node and set the Copilot Grid Configuration as follows:
- After the Copilot: Set Grid node, add an LLM Prompt node. Configure it to generate the next best response for the agent based on the user’s latest input and to store the result in the Input object:
- System Prompt — enter the following:
- In the Storage & Streaming Options section, configure the following:
- How to handle the result — set to Store in Input.
- Input Key to store Result — set to
input.promptResult.
- Add the following Agent Copilot nodes after the LLM Prompt node and configure them as follows:
- Copilot: Next Action Tile node:
- Tile ID —
next-action - Text:
This expression dynamically retrieves the LLM-generated suggestion for the tile. - Tile ID —
- Copilot: Next Action Tile node:
- Copilot: Transcript Tile node:
- Tile ID —
transcript
- Tile ID —
4
Create a Voice Copilot endpoint
- In Deploy > Endpoints, click + New endpoint, select Voice Copilot, and click Create endpoint.
- On the Setup tab, configure the following:
- Endpoint name — enter a unique name, for example,
Voice Agent Copilot Endpoint. - Target type — select Flow.
- Flow — select the Agent Copilot flow you created from the list.
- Endpoint name — enter a unique name, for example,
5
Create a voice endpoint
- In Deploy > Endpoints, click + New endpoint, select Voice Gateway, and click Create endpoint.
- On the Setup tab, configure the following:
- Endpoint name — enter a unique name.
- Target type — select Flow.
- Flow — select the handover flow from the list, in this example,
Handover.
- In the Number configuration card, click Add number and select the phone number you use in your contact center. If the number isn’t in the list, add it first in the telephony settings.
- Click Save.
6
Test Agent Copilot
To test Agent Copilot, follow these steps:
- Open your contact center platform and call the number you assigned to the Voice endpoint. The voice AI Agent answers the call and creates a chat session in the contact center platform.
- Say something to the voice AI Agent, for example,
Thanks, I'll wait for the human agent.. Your voice message is sent as text to the chat in the contact center platform. - Find and select the corresponding chat in the contact center platform.
At the start of a conversation, the
Initial message for AI Copilotreply may appear in the human agent’s chat as if it were sent by the end user. This reply is added automatically to trigger Agent Copilot and is visible only to the human agent. - The Agent Copilot workspace displays a widget with the conversation transcript and a widget with the next action.
- Click the icon to paste the next action text into the chat and send the message. The voice agent responds with the text from the next action widget.