Create an AI Roleplay Manually

The Scenario defines the situation in which the Roleplay takes place, including the context, interaction, and desired outcome. When you create a Roleplay manually, describe the situation, the roles of the learner and AI Persona, what happens during the interaction, and the outcome you want the learner to achieve. Include sufficient detail to help define how the AI Persona should respond and how the learner should apply the required skills during the conversation.

Note:

You can only create an AI Roleplay from the Draft stage of the Whatfix Mirror Dashboard.

Use the following steps to create an AI Roleplay using the Whatfix Mirror Dashboard:

1. On the Whatfix Mirror Dashboard, click AI Roleplay.

   

2. Click + AI Create Roleplay.

   

    

3. Click Create manually.

   

4. Enter a name for the AI Roleplay. For example, Customer Support.

   

    

5. Select the required Modality for the AI Roleplay. This defines the mode of interaction between the AI Persona and learner. You can choose one of the following:

     • Audio roleplay: Learners interact with the AI Persona using voice.

     • Chat roleplay: Learners interact with the AI Persona using text.

   

   Info:

   By default, Whatfix selects Audio roleplay as the modality for the AI Roleplay.

6. Click Create.  

   

7. Under the Scenario tab, enter or paste a detailed description of the Roleplay scenario.

   

8. Click Save.

   

   Note:

   Unsaved changes are not included in the preview. The preview reflects the most recently saved version of the AI Roleplay. Click Save to include all latest changes.

   

9. Click Preview AI Roleplay.

   

After you define and save the scenario, proceed to configure the Persona for the Roleplay.

The Persona defines the AI character that interacts with the learner during the Roleplay. Configure the Persona to establish its identity, characteristics, and behavior based on the Roleplay scenario and learning objectives.

Use the following steps to define the Persona:

1. Click the Persona tab.  

   

2. Click Create persona.    

   

   Info:

   If you want to select an existing persona, click Choose persona and select one of the available personas to proceed. For more information, see Configure a Persona for an AI Roleplay.

   

3. Enter a Name for the persona.  

   

4. Enter the Job Title for the persona.

   

5. Select the required Gender for the persona.

   

6. From the Language dropdown, select the required language. For more information, see Supported Languages and Voices.

   

7. From the Voice dropdown, select the required voice. For more information, see Supported Languages and Voices.

   

   Info:

   To hear a preview of the selected voice, click the Voice Preview icon.

   

    

8. Enter a Behavioural description for the persona. This should include how the persona behaves, what they care about and how they should react.

   

9. Click Create persona.  

   

10. Under Persona, review the configured Persona details and confirm that its identity, characteristics, and behavior align with the Roleplay scenario and learning objectives.

    

    Info:

    To add a different persona, click Change persona. For more information, see Configure a Persona for an AI Roleplay.

    

11. In the Conversation starters section, under Conversation type, click the Open icon.  

    

12. Select the required Roleplay Conversation type. For more information, see Understand the Types of Roleplay Conversations.

    

13. Select who speaks first in the conversation. You can choose from the following options:

      • AI persona: The AI persona starts the conversation.

      • Learner: The learner starts the conversation.

    

14. Enter the appropriate Opening message and Closing message based on whether the AI persona or learner speaks first.

    

After you complete the Persona configuration, proceed to specify the Evaluation parameters for the Roleplay.

Use Evaluation parameters to define what the learner should demonstrate and how their performance should be assessed during the Roleplay. For each parameter, you can specify the evaluation type, competencies, scoring guidance, and reference examples.

The score for each parameter contributes to the learner's overall evaluation. The total score is calculated at the end of the Roleplay based on the learner's performance across the parameters.

Note:

You can add up to 10 evaluation parameters to an AI Roleplay. Ensure that the total number of parameters does not exceed this limit.

Use the following steps to add Evaluation parameters for the Roleplay:

1. Click the Evaluation tab.    

   

2. Click the Open icon.

   

3. Select the required type of evaluation parameter.
   

   Info:

   Whatfix enables you to choose from the following evaluation parameters:  

   Evaluation Parameter	Description
   Rating	Evaluate performance using a predefined numerical scale, such as 0–10. Use this to measure how well the learner demonstrates a specific competency.
   Text	Evaluate performance using written feedback or qualitative observations. Use this when the evaluation requires a descriptive response rather than a numeric score.
   Single Choice	Evaluate performance by selecting one option from a predefined list. Use this when only one outcome or level best describes the learner’s performance.
   Multiple Choice	Evaluate performance by selecting one or more options from a predefined list. Use this when multiple outcomes or behaviors may apply to the learner’s performance.

   Tip:

   Whatfix recommends you to use the Rating scale, as it provides clear, quantifiable feedback on learner performance.

4. Enter a name for the new evaluation parameter. For example, Empathy.

   

5. Click the Competency dropdown to open the competency selection list.

   

6. Select the required competencies for the evaluation parameter by using the corresponding checkboxes.

   

7. Enter the Scoring guidance for the evaluation parameter. Define the level of learner performance that corresponds to each score range. For example, describe what scores of 0–2, 3–5, 6–8, and 9–10 indicate based on the learner’s performance.

   

8. In the Reference examples section, under Good response, enter one or more examples of the expected response.
   Example: “I understand how frustrating this must be. I’ll look into what happened with your order and help get this resolved.”

   

9. In the Reference examples section, under Poor response, enter one or more examples of the expected response.
   Example: “You’ll need to contact billing. There’s nothing I can do about it.”

   

   Info:

   To add additional evaluation parameters, click + Add parameter and repeat the steps above.

   

After you add the required evaluation parameters, proceed to specify the Exit conditions for the Roleplay.

Exit conditions control when and under what circumstances the Persona ends the Roleplay. They help ensure the conversation concludes naturally if the learner doesn’t meet the required scenario objectives or if the interaction reaches its defined limit.

Use the following steps to specify exit conditions for an AI Roleplay:

1. Click the Exit conditions tab.

   

2. Review the exit conditions.

   

3. Click the Open icon.

   

   Note:

   You can configure a Roleplay session to last up to 15 minutes.

4. Enter the Repetition limit for the Roleplay session. This defines how many times the AI persona can repeat the same question before the session ends. You can set this limit up to 5.

   

   Info:

   Enter custom rules that define when the Roleplay session should end. For example, you can end the Roleplay session when the customer asks to speak to a manager.

   

After you configure the Exit conditions, proceed to define the Learner brief for the AI Roleplay.

The Learner brief provides context for the Roleplay by outlining the scenario, learner objective, and expectations. When you create a Roleplay manually, enter a brief that gives the learner the information they need before the conversation starts.

Example: You are a support agent handling a customer frustrated about a delayed order. Listen actively, acknowledge the issue, and provide a clear resolution to de-escalate the situation.

Note:

You can add a Learner brief of up to 500 characters.

Use the following steps to define the Learner brief for an AI Roleplay:

1. Click the Learner info tab.

   

2. Enter the Learner brief.
   

   Tip:

   Keep the Learner brief concise and use the second person to address the learner.

3. Click Save.  

   Note:

   Whatfix does not save your changes automatically. Click Save each time you make a change to the Roleplay to ensure the preview reflects your latest changes.

   

After you review and save the AI Roleplay, proceed to attach it to a Workflow.

Once you create and save an AI Roleplay, you must attach it to a Workflow to integrate the Roleplay with the Mirror experience. Alternatively, you can also attach a Workflow to an AI Roleplay. For more information, see Attach a Roleplay in Mirror.

Info:

Whatfix creates every new AI Roleplay in the Draft stage. Attach the Roleplay to a Workflow before you move it to the Ready or Production stages.

Use the following steps to attach the created AI Roleplay to a Workflow:  

1. Click Attach to workflow.

   

    

2. Select the required stage that contains the AI Roleplay you want to attach. For more information, see Understand Whatfix Dashboard Stages.

   

3. Select the Workflow to which you want to attach the AI Roleplay.

   

4. Click Attach.

   

Note:

Once you attach an AI Roleplay to a Workflow, preview the combined Mirror Roleplay experience. For more information, see Preview a Roleplay in Mirror.