Getting Started with the Async Text-to-Speech Streaming API

Welcome! This quick-start guide walks you through sending your first request and turning text into real-time audio.

Prerequisites

Developer account on Async

API key

A command-line HTTP client (examples use cURL; feel free to use Postman, Python requests, etc.)

ffmpeg (optional, but handy for converting the raw stream to other formats for playback)

Navigate to API Keys → Create API Key.

Copy the key and store it securely (it starts with sk_).

You can export it as an environment variable so it never appears in your shell history:

Here’s a simple curl command that sends text and receives streamed audio in response:

The command above stores 44.1 kHz 16-bit mono PCM samples in speech.pcm.

Quick playback (macOS/Linux)

Convert to WAV

Tip: For real-time playback, pipe the cURL output directly into ffplay:

HTTP Code	Meaning	Most common fixes
401 Unauthorized	Bad or missing X-Api-Key	Check key spelling; confirm it has TTS scope.
429 Too Many Requests	You hit the rate limit.	Wait for the mentioned time and retry (or upgrade plan).
400 Bad Request	Validation error in your JSON.	Validate JSON, field names, and ranges.

Need help? Ping us or hop into our developer Discord. Happy building!