All termsGlossary

What is natural language understanding (NLU)?

Natural language understanding works out what a piece of text means. It identifies what the person wants and picks out details such as dates, names and order numbers.

Why it matters

People phrase the same request in many ways, and they rarely use the words on a menu. Understanding lets a system respond to the meaning of a sentence instead of matching fixed keywords. When it works, callers can explain a problem in their own words and still reach the right answer or team. When it fails, they are asked to repeat themselves or are sent to the wrong place.

How it works

The input is text, either typed by the customer or produced by speech recognition. A model classifies the text into an intent, such as cancelling an order, and extracts entities, the specific values inside the sentence. It also uses earlier turns of the conversation to resolve references such as “that one”. Older systems were trained on example phrases for each intent. Newer ones often use a large language model.

In practice

Speech recognition passes on the text “I was charged twice for last month.” Understanding labels the intent as a billing dispute and extracts two details: a duplicate charge and the period, last month. The system can then look up the account’s recent payments and ask a useful follow-up question, instead of reading out a general billing menu.

Get Started

Let’s build AI agents, together.

Talk with us to see the top three use cases for AI agents at your business, and how quickly you could go live.

Live in 7 days · Works inside the tools your team already uses · Hands off with full context