Opal Logo
WhatsApp

Avatar by Opal Interactive

An interactive brand avatar your audience can talk to.

Avatar by Opal Interactive is a real-time host for websites, brand activations, museums and trade shows. It listens, answers from your own content without breaking character, and runs in the browser: no render farm, no per-minute video bill.

Universidad Cenfotec Avatar by Opal Interactive × SpatialLab
Opal avatar: a live screenshot of Curio, a friendly brown-haired character in a teal shirt, over the demo’s animated lavender brand background

Web avatar

Why Avatar

  • A character, not a chatbot

    Lipsync, expressions and gaze tracking make it read as someone present, not a text box with a face pasted on. That is what makes people stop, talk and stay.

  • Grounded in what you actually know

    It answers from your own knowledge base, not a generic script. When it does not know something, it says so and offers what it does know, instead of inventing an answer about your business.

  • Runs in the browser, not the cloud

    The character renders on your visitor’s own device: zero render cost, and it keeps working on-prem, at the edge or fully offline.

Web avatar

Use cases

24/7 virtual advisor

A branded presence on your site that answers real questions, day or night, without a queue.

Trade show and lobby kiosks

Push-to-talk, full screen, no keyboard: built for a walk-up audience with fifteen seconds of attention.

Museums and exhibitions

A guide that answers what each visitor actually wants to know, at their own pace.

Private, on-prem deployments

For teams that cannot send visitor conversations to a third-party cloud, it runs fully local.

Web avatar

How it works

You bring the brand and the knowledge. We handle the character, the model and the tuning behind it.

  1. Character and voice

    We design the character with you: look, voice, personality and tone, so it represents your brand instead of a generic assistant.

  2. Knowledge base

    We load what it should know from your real content and define what it must not say. Every answer stays grounded in that base.

  3. Deployment

    It goes live as an embed on your site or as a full-screen, push-to-talk kiosk at your activation, booth or gallery.

  4. Tuning on real conversations

    We review what people actually ask and tighten the answers, so the avatar gets better at its job the longer it runs.

Web avatar

See it live

The best way to understand Avatar is to talk to it.

Open the full experience

Origin

Born in a research lab, built for the market

The technology behind Avatar was born at SpatialLab, the applied R&D lab of Universidad Cenfotec. There the VTuber idea evolved into something more useful, a real-time conversational host for brands, and the lab connected that research with industry.

Opal Interactive develops that base as a product: character craft, grounded conversation, deployment and support. Research explores what is possible; we turn it into something your audience actually uses.

Universidad Cenfotec

Avatar by Opal Interactive × SpatialLab

Learn more about SpatialLab

Web avatar

Technical details

Rendering
Real-time character rigs in the browser (WebGL), no per-minute video cost
Conversation
Voice and text, streamed sentence by sentence, interruptible mid-answer
Knowledge
Grounded answers from your approved knowledge base, pluggable AI brain (cloud or local)
Presence
Viseme lipsync, expressions and webcam gaze tracking
Deployment
Web embed or full-screen kiosk; cloud, on-prem or fully offline

Web avatar

Frequently asked questions

  • How is Avatar different from a chatbot?

    A chatbot is a text box. Avatar by Opal Interactive is an embodied host: a character with a voice, expressions and gaze that reads as someone present. That difference is what makes people stop and talk, on a website and even more so in a physical space.

  • How does it know what to answer?

    It answers from a knowledge base built from your real content, in a tone defined with you. When a question falls outside that base, it admits it and offers what it does know instead of making something up.

  • Where can it run?

    Anywhere a modern browser runs: your website, a full-screen kiosk at a fair or a museum, a lobby screen. It can run in the cloud, on-premise or fully offline, which matters for teams that cannot send conversations to third parties.

  • Do we pay per minute of video?

    No. Unlike generated-video avatars, the character renders on the visitor’s own device, so there is no render farm and no per-minute bill. The character rig is an asset you own.

Web avatar

Put a host in front of your audience

Tell us where your audience meets you, on the web or in a physical space, and we will propose what your brand’s avatar could do there.