# Listen

> Waits until the caller finishes talking, then saves their words for the next step.

Source: https://voxio.in/docs/steps-listen · voxioagent 2.0.0 · Index: https://voxio.in/docs/llms.txt

Type: `input`.

## What it does

1. Pauses this line of the workflow until the caller finishes speaking.
2. Saves their words under each name in `input_variables`, normally `user_input`. Values from the last turn are cleared first, so a step after Listen never reads an old answer.
3. Carries on with its `next`.

Most workflows have one Listen step that every turn comes back to. When the caller counts as “finished” is set on the flow: see [Turn-taking](https://voxio.in/docs/turn-taking.md).

## Settings

Everything goes in the step's `parameters`. In the editor, click the step to open its settings.

- `input_variables` ({ name: type }): In the editor: **Save what they say as**. The name later steps use to read the caller’s answer. Almost always “What the caller said”. Default: `{"user_input":"str"}`.

## Examples

### The usual Listen

Listen, then think and show the words in the transcript at the same time:

```json
{
  "type": "input",
  "parameters": {
    "input_variables": {
      "user_input": "str"
    }
  },
  "next": [
    "think",
    "transcription"
  ]
}
```

## Next steps

- [Steps](https://voxio.in/docs/steps.md): all the other steps.
- [Patterns](https://voxio.in/docs/workflows-patterns.md): see it in a full workflow.