Encyclopedia

Conversational AI

What conversational AI is, its main components, and how modern LLM-based systems differ from older intent-based ones.

Conversational AI is the field of technology that lets computers hold natural-language conversations with people, spanning chatbots, voice assistants and AI agents, typically built on large language models and speech technology.

The main components

  • Natural-language understanding: interpreting what the user means, not just the literal words.
  • A language model that generates responses and, in agentic systems, decides which actions to take.
  • Dialogue management: keeping track of context across turns.
  • For voice: speech-to-text and text-to-speech to turn audio into text and back.

Intent-based versus LLM-based

Older conversational AI was intent-based: designers defined a fixed set of intents and example phrases, and the system matched each message to one intent and returned a scripted response. It broke on anything unexpected.

Modern conversational AI is built on large language models. Instead of matching to predefined intents, the model understands free-form language and can reason about the request. Combined with tool-calling, it becomes an AI agent that completes tasks, not just answers.

Where it is used

In customer contact, conversational AI powers chat, email, WhatsApp and phone: answering questions, resolving requests and routing to people. The same underlying technology also drives voice assistants and in-product help.

From reference to practice

Turn what you just read into a working AI agent on your own customer contact. Start free, no credit card needed.

    Conversational AI | Conveya