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

# Text to Speech

> Turn text into a spoken audio file with OpenAI.

**Text to Speech** generates spoken audio from a string. Choose a voice, speed,
model, and file format, then connect the resulting file to storage, email, or
another media action.

## Inputs

| Input                   | Type           | Required | Description                                                                  |
| ----------------------- | -------------- | -------- | ---------------------------------------------------------------------------- |
| **Third-party account** | OpenAI account | Yes      | The API key used for synthesis.                                              |
| **Text**                | String         | Yes      | The content to speak.                                                        |
| **Model**               | Enum           | Yes      | TTS-1, TTS-1 HD, or GPT-4o Mini TTS. Defaults to TTS-1.                      |
| **Voice**               | Enum           | Yes      | Alloy, Ash, Ballad, Coral, Echo, Fable, Onyx, Nova, Sage, Shimmer, or Verse. |
| **Speed**               | Number         | No       | From `0.25×` to `4×`. Defaults to `1×`.                                      |
| **Response Format**     | Enum           | Yes      | MP3, Opus, AAC, FLAC, or PCM. Defaults to MP3.                               |
| **Instructions**        | String         | No       | Delivery direction available only for GPT-4o Mini TTS.                       |

Example instructions:

```text theme={null}
Speak warmly and clearly, with a brief pause after each sentence.
```

## Output

| Output    | Type | Description                                                       |
| --------- | ---- | ----------------------------------------------------------------- |
| **Audio** | File | The generated audio, named from the first part of the input text. |

<Tip>
  Use TTS-1 for a general-purpose result, TTS-1 HD when fidelity matters more,
  or GPT-4o Mini TTS when you need explicit delivery instructions.
</Tip>