> For clean Markdown of any page, append .md to the page URL.
> For a complete documentation index, see https://docs.resemble.ai/llms.txt.
> For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://docs.resemble.ai/_mcp/server.

# Getting Started

The WebSocket API delivers low-latency text-to-speech streaming. Use it when you need immediate playback, conversational agents, or continuous audio synthesis.

## Key Features

* **Real-time audio** – Receive PCM or MP3 chunks as they are generated.
* **Flexible payloads** – Choose JSON frames (base64) or raw binary streams.
* **Connection-aware errors** – Separate recoverable synthesis errors from unrecoverable connection failures.

Before you stream:

1. Confirm your account has Business-plan WebSocket access.
2. Retrieve your API token (`Authorization: Bearer ...`).
3. Select an eligible `voice_uuid`.

Continue with [Receiving Audio](./receiving-audio) for request parameters and response shapes.