### Answer Sentiment Analysis in Retell AI detects the emotional tone in a speaker's voice or language and classifies it into three categories: positive, neutral, or negative. [(1) ]() Positive sentiment represents interactions or responses where the emotional tone is favorable, satisfied, or enthusiastic. [(2) ]() Neutral sentiment represents interactions where the emotional tone is neither particularly positive nor negative—typically matter-of-fact or emotionally balanced. [(1) ]() Negative sentiment represents interactions where the emotional tone is unfavorable, frustrated, or dissatisfied. [(2) ]() Retell AI measures sentiment for both users and agents, tracking metrics like User Positive Sentiment Rate, User Negative Sentiment Rate, and Agent Positive Sentiment Rate. The system analyzes both what's said (transcript) and how it's said (tone, pitch, pacing) to classify the overall mood at each stage of the call. [(1) ]()

​ To learn more, see the sources I considered: [[1] Sentiment Analysis](https://www.retellai.com/glossary/sentiment-analysis) [[2] AI QA Metrics](https://docs.retellai.com/ai-qa/terminologies) ​

Feel free to tag @AI Evy with additional questions. I also appreciate feedback, it helps me improve.