# What is turn-taking in a voice agent?

> Turn-taking is how a voice agent decides when the caller has finished, when to reply, and what to do when the caller talks over it.

Source: https://learn.ynteractive.com/content/glossary/turn-taking · Updated 2026-10-21 · Free from Stepthrough (https://learn.ynteractive.com)

[Voice agents](https://learn.ynteractive.com/content/glossary#topic-voice-agents) · [AI glossary](https://learn.ynteractive.com/content/glossary)

Quick answer

Turn-taking is how a voice agent manages who speaks when on a call. It has to guess when the caller has finished, not just paused, answer without a long gap, and stop or keep going when the caller interrupts. Callers who talk over the agent are said to "barge in".

Last updated October 21, 2026

[Take the free module →](https://learn.ynteractive.com/courses/ai-agents-101/6)Module 6 of the free [AI Agents 101](https://learn.ynteractive.com/courses/ai-agents-101) course, about 9 minutes. No account needed.

Updated October 21, 2026 · By [Robert Breen](https://learn.ynteractive.com/content/about)

## Why it matters for a small business

Turn-taking is the difference between a voice agent that feels natural and one that feels broken. Reply too early and it cuts people off mid-sentence, often while they read out a phone number. Reply too late and the caller thinks the line dropped. Ignore interruptions and a caller can't correct a mistake without waiting through a long answer.

Platforms such as ElevenLabs and Vapi handle the detection for you, with settings for how long the agent waits in silence and whether callers can interrupt. Your part is in the prompt and the test calls: short answers, one question at a time, and checking the transcript for places where the agent and caller talked over each other.

## In a real lesson: Your First Voice Agent in ElevenLabs: An After-Hours Receptionist

In [Your First Voice Agent in ElevenLabs](https://learn.ynteractive.com/content/elevenlabs-voice-agent-tutorial-after-hours-receptionist), the First message box explains that if it is empty, the agent waits for the caller to start. Under it sits an Interruptible switch, turned on. Riley's system prompt also says to keep answers short, one or two sentences, like a real phone call, which keeps each turn quick to interrupt or answer.

The lesson doesn't use the term turn-taking; your test call is played with Say it buttons. [The Hartwell Accounting receptionist](https://learn.ynteractive.com/content/ai-receptionist-for-accounting-firm-after-hours-calls) shows why it matters: Casey's message flow asks one thing at a time, yet Dana Whitfield gives her name, number and deadline in one turn, and Casey still reads the number back and marks it urgent.

[Try this lesson free](https://learn.ynteractive.com/modules/elevenlabs-first-agent) or [read the step-by-step guide](https://learn.ynteractive.com/content/elevenlabs-voice-agent-tutorial-after-hours-receptionist).

## Common confusions

### Turn-taking vs latency

Latency is how long the reply takes once the agent decides to speak. Turn-taking is deciding when the caller is done and whose turn it is.

### Turn-taking vs call routing

Call routing sends a call to the right person or line. Turn-taking happens inside one conversation.

## Tips

- Keep the agent's answers to one or two sentences so callers rarely need to interrupt.
- Have it ask one question at a time and read numbers back.
- On test calls, pause mid-sentence and interrupt on purpose, then read the transcript.

## Related terms

[Voice agent](https://learn.ynteractive.com/content/glossary/voice-agent) · [Speech-to-text](https://learn.ynteractive.com/content/glossary/speech-to-text) · [AI latency](https://learn.ynteractive.com/content/glossary/ai-latency) · [Text-to-speech](https://learn.ynteractive.com/content/glossary/text-to-speech) · [Conversational AI](https://learn.ynteractive.com/content/glossary/conversational-ai) · [Multi-turn conversation](https://learn.ynteractive.com/content/glossary/multi-turn-conversation)

## More Voice agents terms

[Virtual receptionist](https://learn.ynteractive.com/content/glossary/virtual-receptionist) · [Voice cloning](https://learn.ynteractive.com/content/glossary/voice-cloning) · [Voicemail transcription](https://learn.ynteractive.com/content/glossary/voicemail-transcription) · [Word error rate (WER)](https://learn.ynteractive.com/content/glossary/word-error-rate)

## Where you use it: free lessons

- [Your First Voice Agent in ElevenLabs: An After-Hours Receptionist](https://learn.ynteractive.com/content/elevenlabs-voice-agent-tutorial-after-hours-receptionist) (ElevenLabs Agents, 9 min)
- [Never Miss an After-Hours Client Call: An AI Receptionist for a Small Office (ElevenLabs)](https://learn.ynteractive.com/content/ai-receptionist-for-accounting-firm-after-hours-calls) (ElevenLabs Agents, 11 min)
- [ElevenLabs vs Vapi: Cost per Call and a Launch Checklist](https://learn.ynteractive.com/content/elevenlabs-vs-vapi-cost-per-call-launch-checklist) (ChatGPT, 8 min)

## Frequently asked questions

**What does barge-in mean?**

The caller starts talking while the agent is still speaking. A good agent stops, listens and responds to what they said.

**Why does my voice agent cut people off?**

It decided a short pause meant the caller had finished. Platforms let you adjust how long it waits; shorter questions also help.

**Should callers be able to interrupt the greeting?**

Usually yes, so returning callers can get to the point, as long as the AI disclosure is short and comes first.

[All AI glossary terms, A to Z](https://learn.ynteractive.com/content/glossary) · [Free prompt templates](https://learn.ynteractive.com/content/prompts)
