"My name is... my name is...": A Linguistic Map for Voice Agents — Midam Kim, ServiceNow

Voice AI systems frequently fail because they treat communication as a rigid technical pipeline rather than a dynamic, joint activity. To improve user satisfaction, developers must adopt a linguistic framework that accounts for the interdependence of sound, word choice, interaction timing, and mental model alignment. Current failures often stem from poor turn-taking, lack of context retention, and an inability to handle emotional nuances. By implementing rigorous benchmarks like EVABench and prioritizing the cognitive experience of the user, organizations can significantly reduce task failures and live agent escalations. Ultimately, voice AI must evolve to mirror human communication, where both parties continuously adapt to each other’s speaking styles and intent over the course of a conversation. Success requires viewing voice interaction as a cognitive, timeline-based process rather than a series of isolated data exchanges.