Sentiment analysis
Sentiment analysis is the automatic assessment of the emotional tone of speech or text: positive, neutral or negative, and often finer states such as frustration or urgency. It reads the words used and, in voice, how they were said.
Also called: tone analysis, emotion detection
On a call, sentiment can be read from vocabulary ("this is the third time I’ve rung"), from phrasing, and from acoustic cues such as pace, volume and pitch. Analysis can run during the call, to change how the agent responds or to trigger an escalation, or afterwards, to tag the call for review.
Its value to a business is triage. A hundred calls a week cannot all be listened to; a sentiment score surfaces the ones that went badly, the callers who were upset, and the patterns behind them. It is also a fair proxy for whether an automated receptionist is actually acceptable to callers, measured on real calls rather than on a demo.
It is an estimate, not a verdict. Sarcasm, cultural differences in expressiveness and background stress all confuse it, and a score should prompt a person to listen rather than replace their listening.
Related
aiReceptionist does this: Call sentiment