Creatium APIs

Voice, avatar, and video APIs

You’ll save 6–10× on cost against comparable alternatives, with the same models behind Creatium’s own interactive products, in production form.

Coming in September.

Example request

hear a voice in one call

curl https://api.creatium.com/v1/voice/speak \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"voice": "coach-en-2", "text": "Run that answer again."}'

response

{
"audio_url": "https://cdn.creatium.com/voice/8f2c1a.mp3",
"duration_ms": 2140
}

Voice

/v1/voice
Natural, controllable speech you’ll be able to tune to your brand.

POST /v1/voice/speak
{ "voice": "coach-en-2", "text": "Go again." }

→

{ "audio_url": "…8f2c1a.mp3", "ms": 2140 }

Avatar

6× lower cost than comparable alternatives, with visuals integrated.

Video

Realistic films for training, and you’ll be able to make them interactive.

Quality

In our internal testing, Creatium Voice produces 2 to 3 times fewer pronunciation errors, with a customizable pronunciation dictionary and natural, controllable speech.

Pricing

It’ll be usage-based, billed separately from the coaching plans. Tell us what you’re building and the volume you expect, and we’ll scope it with you.

How is API usage billed?

The Voice, Avatar, and Video APIs will be usage-based, and billed separately from Coach's own plans. Tell us what you're building and the volume you expect, and we'll scope the pricing with you.

How do I get access?

API access isn't open yet: not self-serve, and not by request; there are no keys to hand out. It opens in September. Sign up for updates and we'll email you when it does.

What about security and subprocessors?

We do not use any input, output, or personal information to train third-party AI. The APIs will run on the same disclosed subprocessors as the rest of Creatium: Anthropic, OpenAI, ElevenLabs, Cartesia, Perplexity, AWS, and Microsoft Azure. See the Security page for the full posture.