Important points from week 3:

1. What is personality? What are the different personality traits? (Big 5)
2. How can personality be expressed? ( linguistic style, voice, gesture..)
3. Complementary vs similarity attraction theory 
4. Other-Personality adaptive chatbots
5. Personas of AI users

Questions for the experiment:

1. How do different embodiments/avatars affect our perception of the LLM?
2. How does the LLM adapts to our personality (perceives our personality and copies) and emotions?
3. How does the LLM persuade us that it understands and cares?
4. How does the embodiment and adaptability of the LLM affect our trust and autonomy?

In the experiment, I want the user to have a seamless conversation with the LLM, as they would in their daily life with ChatGPT. However, I would like to break the illusion that the LLM understands their emotions and show its true colours. Show that it is just a computer program with no personality, emotions or agency. Additionally, I want to understand the user’s perception of the AI avatars.

Persona Development

I developed 3 personas based on research findings.

https://www.canva.com/design/DAGic0lLjSo/H78l4LX1cHM1HBVOaJLdOA/edit?utm_content=DAGic0lLjSo&utm_campaign=designshare&utm_medium=link2&utm_source=sharebutton

Resources

NVIVO SOFTWARE : qualitative research

• text to speech + emotions : https://platform.hume.ai/tts/playground
• voices for D-ID :https://speech.microsoft.com/portal/voicegallery
• FACE ANIMATION : https://docs.d-id.com/reference/talks-overview

https://github.com/comfyanonymous/ComfyUI