Quickstart

This guide uses ordinary HTTP requests. There is no Audixa SDK package to install: use native fetch, Python requests, or cURL.

Step 1: Prepare Your Account

Preview the Voices

Listen to fixed samples at audixa.ai/voices. The sampler does not accept custom text or create free generations.

Enable Paid Generation

Create an account, then purchase at least a $10 top-up or start a subscription. Audixa does not issue trial credits or free generations.

Create an API Key

Open Developers → API Keys, create a key, and copy it when it is shown.

Keep your API key secret. Do not place it in browser code shipped to end users, public repositories, logs, or support messages.

Step 2: Choose a Runtime

Browser console or Node.js 18+ ES module (native fetch)

python -m pip install requests

A POSIX shell with cURL and jq

Step 3: Create and Poll a Generation

POST /v3/tts starts an asynchronous job. The request text must contain at least 1 character, and the maximum length depends on the active plan. Each generation has a minimum billable length of 30 characters (8 tokens), so shorter sentences are accepted and charged at that minimum. Poll GET /v3/tts?generation_id=... until the status is COMPLETED or FAILED.

const API_KEY = 'YOUR_API_KEY';
const API_URL = 'https://api.audixa.ai/v3/tts';
const headers = {
  'X-API-Key': API_KEY,
  'Content-Type': 'application/json'
};

const createResponse = await fetch(API_URL, {
  method: 'POST',
  headers,
  body: JSON.stringify({
    text: 'Hello from Audixa. This example is long enough to generate speech.',
    voice_id: 'am_ethan',
    model: 'base',
    audio_format: 'mp3'
  })
});

if (!createResponse.ok) throw new Error(await createResponse.text());
let generation = await createResponse.json();

while (!['COMPLETED', 'FAILED'].includes(generation.status)) {
  await new Promise((resolve) => setTimeout(resolve, 2000));
  const statusResponse = await fetch(
    `${API_URL}?generation_id=${encodeURIComponent(generation.generation_id)}`,
    { method: 'GET', headers: { 'X-API-Key': API_KEY } }
  );
  if (!statusResponse.ok) throw new Error(await statusResponse.text());
  generation = await statusResponse.json();
}

if (generation.status === 'FAILED') {
  throw new Error(generation.error_message || 'Generation failed');
}

console.log(generation.audio_url);

import time
import requests

API_KEY = "YOUR_API_KEY"
API_URL = "https://api.audixa.ai/v3/tts"
headers = {"X-API-Key": API_KEY}

response = requests.post(
    API_URL,
    headers=headers,
    json={
        "text": "Hello from Audixa. This example is long enough to generate speech.",
        "voice_id": "am_ethan",
        "model": "base",
        "audio_format": "mp3",
    },
    timeout=30,
)
response.raise_for_status()
generation = response.json()

while generation["status"] not in {"COMPLETED", "FAILED"}:
    time.sleep(2)
    response = requests.get(
        API_URL,
        headers=headers,
        params={"generation_id": generation["generation_id"]},
        timeout=30,
    )
    response.raise_for_status()
    generation = response.json()

if generation["status"] == "FAILED":
    raise RuntimeError(generation.get("error_message") or "Generation failed")

print(generation["audio_url"])

#!/usr/bin/env sh
set -eu

API_KEY="YOUR_API_KEY"
API_URL="https://api.audixa.ai/v3/tts"

generation=$(curl --fail-with-body --silent --show-error \
  -X POST "$API_URL" \
  -H "X-API-Key: $API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "text": "Hello from Audixa. This example is long enough to generate speech.",
    "voice_id": "am_ethan",
    "model": "base",
    "audio_format": "mp3"
  }')

generation_id=$(printf '%s' "$generation" | jq -r '.generation_id')
status=$(printf '%s' "$generation" | jq -r '.status')

while [ "$status" != "COMPLETED" ] && [ "$status" != "FAILED" ]; do
  sleep 2
  generation=$(curl --fail-with-body --silent --show-error --get \
    "$API_URL" \
    -H "X-API-Key: $API_KEY" \
    --data-urlencode "generation_id=$generation_id")
  status=$(printf '%s' "$generation" | jq -r '.status')
done

if [ "$status" = "FAILED" ]; then
  printf '%s\n' "$generation" | jq -r '.error_message // "Generation failed"' >&2
  exit 1
fi

printf '%s\n' "$generation" | jq -r '.audio_url'

{
  "generation_id": "gen_abc123xyz789",
  "status": "COMPLETED",
  "voice_id": "am_ethan",
  "model": "base",
  "audio_url": "https://cdn.audixa.ai/audio/gen_abc123xyz789.mp3",
  "error_message": null
}

Request Controls

Field	Requirement
`text`	Required; at least 1 character; minimum charge is 30 characters (8 tokens); maximum length is set by the active plan
`voice_id`	Required; must exist, be accessible, and match the selected model
`model`	`base` or `advanced`; defaults to `base`
`audio_format`	`wav` or `mp3`; defaults to `wav`
`speed`	Optional delivery multiplier from `0.5` to `2.0`
`cfg_weight`	Advanced model guidance control from `1.0` to `5.0`
`exaggeration`	Advanced model expressiveness control from `0.0` to `1.0`
`language_code`	Optional; accepted values depend on the model

Next Steps

Python REST Guide

Structure the same workflow with Python requests.

JavaScript REST Guide

Use native fetch with explicit status polling.

Explore Voices

List the voice IDs available to your account.

API Reference

Review every accepted request and response field.

Getting Started

API Guides

Language Examples

Dashboard

Step 1: Prepare Your Account

Step 2: Choose a Runtime

Step 3: Create and Poll a Generation

Request Controls

Next Steps

Python REST Guide

JavaScript REST Guide

Explore Voices

API Reference

​Step 1: Prepare Your Account

​Step 2: Choose a Runtime

​Step 3: Create and Poll a Generation

​Request Controls

​Next Steps

Python REST Guide

JavaScript REST Guide

Explore Voices

API Reference

Step 1: Prepare Your Account

Step 2: Choose a Runtime

Step 3: Create and Poll a Generation

Request Controls

Next Steps