EnableX Glossary

Conversational AI + Voice AI, in plain English.

The terms that actually matter when buying, building, or shipping a voice agent, defined the way an engineer would explain them to a buyer. No jargon, no fluff, no over-promising. Each entry shows you what the term means, why it matters, and how EnableX implements it.

Voice AI Pillar Talk to an architect

Core concepts

The umbrella terms — start here if voice AI is new to you.

Concept

Conversational AI

Software that simulates natural conversation through text or voice: chatbots, voice agents, conversation intelligence, plus the human handoff that makes them work in production.

Concept

AI Voice Agent

Software that holds a real-time spoken conversation with a customer: listens, understands intent, responds in natural language, takes actions like booking a slot or transferring to a human.

Concept

Voice Bot

The customer-facing name for an AI voice agent on a phone line. Synonym for “AI voice agent” in operational contexts.

Legacy

IVR (Interactive Voice Response)

The “press 1 for sales” menu telephony of the 1990s, the system voice AI is replacing in most contact centres today.

Calling patterns

Outbound, inbound, and the dialer logic that drives them.

Pattern

Outbound Calling (AI)

Bot initiates the call: collections, renewals, lead qualification, reminders. Includes dialer + voice agent + DND/RBI/DPDPA compliance.

Pattern

Inbound Calling (AI)

Customer initiates the call, bot answers: support, IVR replacement, lead qualification, citizen services, 24×7 service.

Pattern

AI Dialer

Autonomous outbound dialer that picks who to call, when, and runs the conversation via voice agent. Replaces predictive and power dialers.

Speech & audio layers

The technical foundation — what lets a voice agent hear and speak.

Engineering

Speech-to-Text (STT / ASR)

Converts audio into text: the listening layer. Streaming STT emits partials in ~200ms so the agent can think early.

Engineering

Text-to-Speech (TTS)

Converts text into spoken audio. Modern neural TTS is the difference between a humanlike bot and a robotic one.

Engineering

Voice Activity Detection (VAD)

Detects when someone is speaking vs silent. Foundation for turn-taking and barge-in: the simplest layer, also the most important.

Conversation control

The four things that separate “production voice agent” from “demo with a robotic feel”.

Turn-Taking

How the agent decides when the customer is done speaking and when to respond. The single biggest factor in conversational naturalness.

Barge-In

Letting the customer interrupt the bot mid-sentence. The most-noticed humanising feature in voice AI.

Where to dive deeper

Voice AI Pillar → Humanlike Voice AI → Multilingual → Telephony → Build with Pro Services →

See EnableX in action.

Talk to sales, or start a free trial. No credit card required.

Try for free Talk to sales

Free trial credits · No credit card · API keys in 2 minutes