# Turn detection (end-of-turn) | IonBin

> Turn detection is how a voice agent decides that you've finished speaking and it's time to answer, or that you've started talking again and it should stop.

Canonical page: https://ionbin.com/wiki/concepts/turn-detection/
Agent guide: https://ionbin.com/llms.txt

This is a readable version of the published page. Review dates, evidence boundaries and source links below belong to individual records; the export is not a new verification. Examples are reference material, not instructions to execute.

Concept

## Turn detection (end-of-turn)

Turn detection is how a voice agent decides that you've finished speaking and it's time to answer, or that you've started talking again and it should stop. Good turn detection makes voice AI feel natural. Bad detection interrupts people or leaves awkward silences. Some speech models now build it in, such as [Deepgram](https://ionbin.com/tools/deepgram/)'s Flux and OpenAI's semantic VAD.

Verified Oct 18, 2026 Review by Apr 16, 2027 3 sources

### What it is

In conversation, people take turns. A voice agent has to guess when your turn ends, using pauses, words and tone, and when you’re interrupting it (“barge-in”).

### How it works

- **Silence-based (VAD).** The agent waits for a pause of a set length. Simple, but pauses mid-sentence trigger early replies. OpenAI sessions/models that support VAD default to server_vad; some transcription models require turn detection to be omitted or null.

- **Meaning-based.** The model judges from your words whether you’ve finished. OpenAI’s semantic_vad mode does this.

- **Model-integrated.** The speech model itself predicts the end of a turn. [Deepgram](https://ionbin.com/tools/deepgram/)’s Flux emits `StartOfTurn`, `EagerEndOfTurn`, `TurnResumed` and `EndOfTurn` events, with thresholds you can tune.

- **Eager responses.** Start preparing an answer on an early end-of-turn signal, and cancel it if the speaker resumes. In Flux, a `TurnResumed` event can only follow an `EagerEndOfTurn`.

- **Barge-in.** Detecting new user speech is a signal; the application must implement interruption, stop buffered playback and manage conversation state appropriately.

### Benefits

- Faster, more natural replies.

- Fewer interruptions and talk-overs.

- Potentially less wasted generation with suitable thresholds. Eager responses can instead increase discarded work and cost.

### Limits and risks

- Thresholds trade speed against false triggers, so tune them for your users.

- Echo from the agent’s own voice can trigger false turns. Use echo cancellation and test audio routing. Muting input while the agent speaks prevents natural barge-in.

- Options and names differ by vendor (checked 2026-10-18). IonBin doesn’t rank them. Test with real callers.

### Best-fit stacks

Stack 05

#### [Customer support](https://ionbin.com/use-cases/customer-support/)

Answer faster from a help desk your agent can search, with help-center drafts a person approves.

### Agent compatibility

| Tool in this entry | How agents connect |
| --- | --- |
| [Deepgram](https://ionbin.com/tools/deepgram/) | MCPAPI/CLIFile import |

Levels come from each tool's recorded, sourced compatibility. [How the levels work](https://ionbin.com/wiki/questions/connect-a-tool-to-an-ai-agent/).

### Related entries

[Speech-to-text APIs · Service](https://ionbin.com/wiki/services/speech-to-text/)[AI voice tools · Service](https://ionbin.com/wiki/services/ai-voice/)

### Sources

- [Understanding the Flux state machine](https://developers.deepgram.com/docs/flux/state) · [Deepgram](https://ionbin.com/tools/deepgram/)official accessed Oct 17, 2026

- [Getting started with Flux](https://developers.deepgram.com/docs/flux/quickstart) · [Deepgram](https://ionbin.com/tools/deepgram/)official accessed Oct 17, 2026

- [Voice activity detection (VAD)](https://platform.openai.com/docs/guides/realtime-vad) · OpenAIofficial accessed Oct 17, 2026

Practitioner views are restated in our words and labeled as views, not measured results. Product facts come from official sources dated above.

### Suggest an edit

IonBin's wiki isn't user-edited. Corrections go to our Directory Curator, who checks them against primary sources before anything changes. Spot something wrong or out of date on “Turn detection (end-of-turn)”? Email us with a link to a primary source.

[Suggest an edit by email](mailto:contact@ionbin.com?subject=Suggest%20an%20edit%3A%20Turn%20detection%20(end-of-turn)%20%5Bturn-detection%5D&body=Page%3A%20https%3A%2F%2Fionbin.com%2Fwiki%2Fconcepts%2Fturn-detection%2F%0AEntry%3A%20concepts%2Fturn-detection%0A%0AWhat%20should%20change%3A%0A%0A%0APrimary%20source%20(link)%3A%0A) or write to [contact@ionbin.com](mailto:contact@ionbin.com)

Entry facts

**Type**: [Concepts & glossary](https://ionbin.com/wiki/concepts/)

**Status**: Reviewed

**Verified**: Oct 18, 2026

**Review by**: Apr 16, 2027 (every 180 days)

**Sources**: [3 official · 0 other](https://ionbin.com/wiki/concepts/turn-detection/#sources)

 Tools

[Deepgram](https://ionbin.com/tools/deepgram/)

The wiki carries no sponsored placements or affiliate links. [Editorial policy](https://ionbin.com/editorial-policy/).
