> For the complete documentation index, see [llms.txt](https://frostember-studios.gitbook.io/frostspeech-tts/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://frostember-studios.gitbook.io/frostspeech-tts/getting-started/quickstart.md).

# Core Concepts & Engines

FrostSpeech TTS supports multiple Text-to-Speech engines, ranging from fully local/offline generation to premium cloud voices. Understanding their differences is important for choosing the right workflow for your project.

**Piper TTS (Local/Offline):**\
An extremely fast local neural TTS engine. It uses `.onnx` voice models and relies on eSpeak-NG for text-to-phoneme conversion. Piper is completely free to use after setup, runs locally on your machine, and is ideal for quickly generating large amounts of placeholder dialogue, prototype narration, NPC barks, or iteration-heavy voice lines without paying per request.

**Sherpa-ONNX (Local/Offline):**\
A powerful local inference engine for modern speech models. FrostSpeech uses Sherpa-ONNX to run several model families locally, including VITS-style models and Matcha models. It is a good choice when you want higher-quality local generation while still avoiding cloud APIs and recurring generation costs. Like Piper, it runs on your local machine once the required binaries and models are downloaded.

**Kokoro TTS (Local/Offline):**\
A high-quality neural voice option powered through Sherpa-ONNX. Kokoro models are designed for more natural, expressive speech and can produce results closer to premium cloud voices while still running locally. FrostSpeech exposes Kokoro as its own engine so you can select it directly when you want the best local voice quality available in the tool.

**ElevenLabs (Cloud API):**\
A premium cloud-based TTS provider focused on highly realistic and emotional AI voices. This engine requires an active internet connection and your own ElevenLabs API key. Generation may consume ElevenLabs credits depending on your account and usage. It is best suited for final-quality narration, character voices, trailers, prototypes requiring very polished speech, or teams already using ElevenLabs in their production pipeline.

**Edge TTS (Online):**\
An online Microsoft Edge Read Aloud voice option integrated directly into Unity. It does not require a local model download, Python setup, or an API key, but it does require an internet connection and depends on Microsoft’s online voice service availability. It is useful for fast, convenient generation across a large number of languages and voices when you want an online option without setting up a paid cloud account.

In general, use **Piper** for fast local iteration, **Sherpa/Kokoro** for higher-quality local generation, **Edge TTS** for convenient online no-key generation, and **ElevenLabs** when you need premium cloud voice quality.


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter:

```
GET https://frostember-studios.gitbook.io/frostspeech-tts/getting-started/quickstart.md?ask=<question>&goal=<endgoal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.