Empathy & Tone Evaluation for Human-Centered AI Conversations
Focused on evaluating and refining AI-adjacent conversational content with an emphasis on empathy, emotional awareness, and human sensitivity. The project involved reviewing text and audio interactions to identify emotional cues, assess tone appropriateness, and ensure responses aligned with supportive, respectful, and non-judgemental communication standards. Applied psychology-informed judgement and active listening skills to rate conversational quality, flag emotionally misaligned responses, and support the development of AI interactions that feel safe, considerate, and human-centred. Emphasised clarity, ethical sensitivity, and consistency while adhering to structured evaluation guidelines.