Glossary

Intent Recognition

Classification of caller intent into predefined categories (booking, prescription, complaint, info). Classically a classifier; today usually LLM few-shot.

Intent recognition classifies a caller utterance into a predefined intent — "book appointment", "reschedule appointment", "billing question", "complaint". In modern voice-AI stacks the LLM itself usually does this, often combined with a small hardened classifier layer for high-frequency standard cases.

Solid intent recognition needs a reality-based intent inventory: too many intents (>30) make the model uncertain, too few (<5) leave relevant cases unhandled. A two-stage model usually works best: 6–12 top-level intents plus optional slot fields ("for what date?", "which treatment type?").

Operationally critical: each intent class needs a fallback strategy for low-confidence cases, a measured confusion-matrix report from real calls, and a review cycle in which newly appearing concerns are either added as an intent or escalated to a human.

FAQ
How many intents are reasonable?
For most SMB use cases, 6–12 top-level intents plus slot filling is the right order of magnitude. Anything above tends to become unreliable without producing extra value.
Go deeper in the docs

Next step

See BHOMY in a 15-minute demo on a real call example.

🍪

Cookies & Privacy

We use cookies to provide you with the best possible experience on our website. Some of them are technically necessary, others help us improve the website.