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

# Text to Speech

The Text to Speech step lets you create an AI-generated audio snippet from the script you enter. You can choose models from several popular providers, a specific voice, and (when possible) the target language.

![Text to Speech step](https://mintlify.s3-us-west-1.amazonaws.com/respell-docs-v2/_images/steps-reference/text-to-speech.png)

## Options

| Name     | Type       | Description                                                                                                              |
| -------- | ---------- | ------------------------------------------------------------------------------------------------------------------------ |
| Model    | Dropdown   | The AI model you want to use to narrate the script.                                                                      |
| Prompt   | Plain Text | What you want the AI to say. The AI will narrate exactly what you enter.                                                 |
| Voice    | Dropdown   | The voice model you want the AI to speak in. Models often offer several voices with different personalities and accents. |
| Language | Dropdown   | The target language the AI should speak in.                                                                              |

## Outputs

| Name             | Type       | Description              |
| ---------------- | ---------- | ------------------------ |
| Generated Speech | Audio File | The AI-generated speech. |

## Tips

* Try different models and voices! Some models will handle ambiguities like punctuation or pauses better than others, and voices can often differ in quality as well.
* Usually, you'll want to choose a voice that resonates most with the target listeners.