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

# Speechify

> Use the Speechify text-to-speech plugin in a Zero Runtime pipeline.

Speechify is a **text-to-speech** plugin. It turns the LLM's reply into audio the caller
hears using Speechify's SIMBA model family.

## Setup

Set your Speechify API key in the worker environment. Generate a key from the [Speechify console](https://console.sws.speechify.com/):

```bash theme={null}
export SPEECHIFY_API_KEY=<key>
```

## Usage

Import the plugin and pass it to the pipeline's `tts` slot.

```python theme={null}
from zrt.plugins import SpeechifyTTS

tts = SpeechifyTTS()

# Pipeline(tts=tts, ...)
```

## Parameters

| Parameter     | Type          | Default           | Description                                                                                                                                                                                                                                            |
| ------------- | ------------- | ----------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------ |
| `api_key`     | `str \| None` | `None`            | Speechify API key. Falls back to the `SPEECHIFY_API_KEY` environment variable when unset.                                                                                                                                                              |
| `voice_id`    | `str`         | `"kristy"`        | Speechify voice identifier. Browse the full catalogue at the Speechify voice library.                                                                                                                                                                  |
| `model`       | `str`         | `"simba-english"` | SIMBA model variant. `"simba-english"` is the flagship English model with the highest quality, lowest streaming latency, and full SSML support. `"simba-multilingual"` adds cross-language support. `"simba-base"` and `"simba-turbo"` are deprecated. |
| `sample_rate` | `int`         | `24000`           | Output audio sample rate in Hz.                                                                                                                                                                                                                        |

Spoken replies support interruptions. When the caller talks over the agent, the runtime
stops the audio and listens again. See the [plugins overview](/plugins/overview) for how TTS fits into the rest of the pipeline.