You can improve your AI avatar's response quality by updating key profile settings and providing clear training data and instructions. Follow these steps to make sure your avatar gives better, more relevant, and helpful responses to your learners.
Step 1: Set the right Tone and Style
Navigate to AI Avatars > Profile > Behaviour
Go to the Tone & Style.
The AI generates tone and style based on your existing content, but you can fine-tune it to sound exactly how you speak.
Define how your avatar should sound, for example: “Professional, conversational, concise, etc.”
This helps your avatar adopt a consistent and accurate communication style that matches your brand or subject.
Example:
If you choose “Friendly and supportive,” your avatar will use warm, helpful language like:
“Great question! Let me guide you through it.”
Step 2: Add Custom Instructions
Scroll to Custom Instructions section.
Add short and specific directions to guide how your avatar answers.
Example: “Be clear, concise and direct.”You can add multiple instructions.
Step 3: Set Clear Messages
Navigate to:
AI Avatars > Profile > How it Writes
Initial Message: Set how your avatar greets users in a new conversation.
Example: “Hello! How can I assist you today?”
Message on No Answer: Customize what the avatar says if it can’t respond confidently.
Example: “I’m sorry, I don’t have an answer for that right now.”
Step 4: Configure Response Settings
Scroll Down, review and enable relevant options:
Personalize Responses: Remembers previous chats if the user is logged in.
Favor New Content: Prioritizes showing users unseen content.
Recency Bias (optional): Give priority to recently updated content, if needed.
Step 5: Review and Organize Training Data
Navigate to:
AI Avatars > Brain > All
Make sure only relevant, accurate content is marked as Used for training.
Regularly update or remove outdated content.
Was this article helpful?
That’s Great!
Thank you for your feedback
Sorry! We couldn't be helpful
Thank you for your feedback
Feedback sent
We appreciate your effort and will try to fix the article