# Svara TTS

> Text-to-speech API: 80 languages, a curated and growing voice library, zero-shot voice cloning, and real-time streaming (including WebSocket input streaming). OpenAI- and ElevenLabs-compatible. Change one base URL.

## Getting started

- [Introduction](https://platform.kenpathlabs.com/docs.md): Turn text into lifelike speech in 80 languages with one API call.
- [Quickstart](https://platform.kenpathlabs.com/quickstart.md): From API key to spoken audio in under five minutes.
- [Authentication](https://platform.kenpathlabs.com/authentication.md): Two header styles, one key. Use whichever your client already speaks.

## Capabilities

- [Text to speech](https://platform.kenpathlabs.com/text-to-speech.md): The core endpoint: text in, audio out, in any of eight formats.
- [Streaming](https://platform.kenpathlabs.com/streaming.md): First audio in a few hundred milliseconds. Stream by default.
- [Input streaming](https://platform.kenpathlabs.com/input-streaming.md): Pipe an LLM's token stream straight in. Speech starts before the sentence exists.
- [Voice cloning](https://platform.kenpathlabs.com/voice-cloning.md): Clone any voice from a few seconds of reference audio.
- [Voices & languages](https://platform.kenpathlabs.com/voices.md): A curated voice library across 80 languages, plus your own clones.

## Reference

- [Rate limits & errors](https://platform.kenpathlabs.com/rate-limits.md): Plan limits, 429 semantics, and every error shape you can receive.
- [SDKs & compatibility](https://platform.kenpathlabs.com/sdks.md): The official OpenAI and ElevenLabs SDKs work unmodified. Change one URL.
- [API reference](https://platform.kenpathlabs.com/api-reference.md): Every endpoint at a glance.

## Optional

- [Full documentation (one file)](https://platform.kenpathlabs.com/llms-full.txt)