Use a generated key to call the AiVoicee voice generation API from any server-side program. Pass the key in the x-api-key header on every request.

AiVoicee gives you a choice of three underlying generation platforms, selected per-request with the provider field on POST /api/v1/tts. Each has its own voice catalog and its own voice_id namespace — a voice_id from one provider's catalog will not work with another.

genaipro tasks are async (queued, poll for the result). elevenlabs and fishaudio tasks are synchronous — the result is already completed the moment POST /api/v1/tts returns — GET is still there if you want it, but you don't have to poll.

The GenAIPro catalog (10,000+ voices), served from our own database — a background job mirrors it locally and re-syncs daily, so this responds instantly with a real, exact total count rather than proxying the provider live. No API key required. Browse it to find a voice_id to use below. Each voice also includes a preview_url — a short sample clip, so you can let your users listen before picking one. No generation/credits needed to play it, it's a static file.

The other two catalogs work identically (same query params, same response shape, same total/has_more paging) at their own endpoints: GET /api/voices/free-tier (ElevenLabs direct, ~20 voices, no query params needed) and GET /api/voices/fishaudio (50,000+ voices, also takes gender/language/category filters).

Response also includes total (exact count matching your filters) and has_more. To pull the whole catalog, page through with page_size=100 until has_more is false (~117 requests for the full 10,000+ list) — cache the result on your side rather than re-fetching it all on every load.

async function fetchAllVoices() {
  const all = [];
  let page = 0;
  while (true) {
    const res = await fetch(`https://api.aivoicee.com/api/voices?page=${page}&page_size=100`);
    const { data } = await res.json();
    all.push(...data.voices);
    if (!data.has_more) break;
    page++;
  }
  return all; // ~11,000+ voices
}

{
  "voice_id": "y2H4TwIU5I2L0JXOdBeX",
  "name": "Aria",
  "gender": "female",
  "accent": "american",
  "preview_url": "https://files.genaipro.io/previews/y2H4TwIU5I2L0JXOdBeX.mp3",
  "image_url": "https://files.genaipro.io/avatars/y2H4TwIU5I2L0JXOdBeX.jpg"
}

<audio controls src="VOICE.preview_url"></audio>

curl -X POST https://api.aivoicee.com/api/v1/tts \
  -H "x-api-key: YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "text": "Hello from AiVoicee!",
    "voice_id": "VOICE_ID_HERE"
  }'

curl -X POST https://api.aivoicee.com/api/v1/tts \
  -H "x-api-key: YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "text": "Hello from AiVoicee!",
    "voice_id": "FISH_VOICE_ID_HERE",
    "provider": "fishaudio",
    "model_id": "s2.1-pro-free"
  }'

const res = await fetch('https://api.aivoicee.com/api/v1/tts', {
  method: 'POST',
  headers: {
    'x-api-key': 'YOUR_API_KEY',
    'Content-Type': 'application/json',
  },
  body: JSON.stringify({
    text: 'Hello from AiVoicee!',
    voice_id: 'VOICE_ID_HERE',
  }),
});
const { data } = await res.json();
console.log(data.task_id);

import requests

res = requests.post(
    "https://api.aivoicee.com/api/v1/tts",
    headers={"x-api-key": "YOUR_API_KEY"},
    json={"text": "Hello from AiVoicee!", "voice_id": "VOICE_ID_HERE"},
)
task_id = res.json()["data"]["task_id"]
print(task_id)

Response:

{
  "success": true,
  "message": "Task created — poll GET /api/v1/tts/:task_id for status",
  "data": { "task_id": "abc123", "char_count": 21 }
}

Poll every 1–2 seconds until status is completed. The result field then holds the audio URL.

curl https://api.aivoicee.com/api/v1/tts/abc123 \
  -H "x-api-key: YOUR_API_KEY"

Notes: keys are shown once at creation — store them securely on your own server, never in client-side / mobile app code. Disable or delete a key below to revoke access immediately.