> ## Documentation Index
> Fetch the complete documentation index at: https://docs.voxworks.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Silence Tolerance

> Silence Tolerance controls how your agent handles periods of user inactivity. It determines when to prompt users who have gone quiet and when to gracefully end calls that have stalled.

## What is Silence Tolerance?

During a conversation, users may pause for various reasons:

* Thinking about their response
* Distracted by something
* Unsure how to proceed
* Technical issues with the call

Silence Tolerance settings control how long the agent waits before taking action.

***

## Key Timers

| Timer                      | Purpose                              | Typical Range  |
| -------------------------- | ------------------------------------ | -------------- |
| **Silence Response Delay** | Time before prompting a silent user  | 4-16 seconds   |
| **Silence Shutdown Delay** | Time before ending an abandoned call | 30-120 seconds |

***

## How It Works

```text theme={null}
User stops speaking
   Response and Shutdown Delay timers start
      User speaks → Timer resets, conversation continues
      Timer expires → Agent prompts user
          User responds → Conversation continues
          No response → Shutdown timer continues
              User speaks → Timer resets
              Shutdown timer expires → Assistant says it is ending the call because it hasn't heard any response
```

***

The `Silence Tolerance` property on each step controls silence tolerance behavior:

| Setting               | Response Delay | Shutdown Delay   | Use Case               |
| --------------------- | -------------- | ---------------- | ---------------------- |
| **hurried**           | approx. 4s     | approx. 15s      | Quick interactions     |
| **patient**           | approx. 16s    | approx. 60s      | Thoughtful responses   |
| **extremely patient** | approx. 48s    | approx. 180s     | Complex decisions      |
| **infinite**          | No prompting   | No auto-shutdown | User-controlled timing |

***

## When to Use Each Setting

### Hurried (Default)

Best for:

* Simple questions with quick answers
* Confirmations and acknowledgments
* Fast-paced conversations

Example: "Is Tuesday at 2pm okay?"

### Patient

Best for:

* Questions requiring some thought
* When users might need to check something
* Multi-part answers

Example: "What challenges are you facing with your current process?"

### Extremely Patient

Best for:

* When users need to look something up
* Fault finding or troubleshooting scenarios
* When users might need to consult others

Example: "Can you check the serial number on the back of the device?"

### Infinite

Best for:

* Situations where users may need extended time
* When you don't want to interrupt at all
* Calls where users explicitly need to step away

Note: Infinite silence tolerance is still limited by the maximum call duration. See [Settings](/scripts/settings) for call duration configuration.

***

## Silence Prompts

When the response delay expires, the agent prompts the user to re-engage:

```text theme={null}
Silence Response: The assistant attempts to continue the current step naturally
Silence Shutdown: The assistant says that it hasn't been able to hear the user, then goodbye and ends the call.
```

***

## Interaction with Other Settings

Silence tolerance works alongside other dynamics:

| Combined With                  | Effect                                                      |
| ------------------------------ | ----------------------------------------------------------- |
| **Patient response eagerness** | Waits longer before responding + longer before following-up |
| **Deep thinking effort**       | More reasoning time + patient silence handling              |

***

## Best Practices

1. **Configure per step** — Set silence tolerance on each step based on the expected response
2. **Match complexity to tolerance** — Complex questions need higher tolerance
3. **Consider your audience** — Elderly users may need more time
4. **Test realistic scenarios** — Try with actual users, not just ideal cases

***

## Next Steps

* [Response Eagerness](/conversation-dynamics/response-eagerness) — Control how quickly the agent responds
* [Thinking Effort](/conversation-dynamics/thinking-effort) — Adjust LLM reasoning depth
* [Overview](/conversation-dynamics/overview) — See all conversation dynamics