Captivation Inside: add a speaking avatar to your AI

For vendors & developers

An embeddable Avatar widget that integrates seamlessly to your AI solution

Upgrade your AI to have a face & a voice

Talk to Us
Request FREE trial

The avatar UI layer powering the next generation of AI-driven customer experiences

WHAT IS CAPTIVATION INSIDE?

Captivation brings your AI to life

Your chatbot or AI service generates the response.

Captivation takes that response and delivers it through an expressive real-time speaking Avatar

You keep your existing AI, LLM, agents, tools and workflows.

Try it here
A digital screenshot showing a woman with straight black hair and fair skin, wearing a light blue turtleneck, on a black background with the text 'Widget test page'. To the right, there is a chat window with a blue header labeled 'Chat', showing a conversation where the user text bubble says 'Hello!' and the reply says 'Welcome! How can I assist you today?'.

Simple to connect

We provide a Javascript widget for you to include the avatar in your front end, and simple developer instructions for you to connect your main LLM  to our platform via the Captivation Bridge.

Request Free Trial
Your front end Your front end
How can I help?
Show me some options
Absolutely
Captivation JavaScript widget
Aimi SuperAgent
Captivation Bridge Captivation
Bridge
Your LLM
WebSocket
Captivation platform Captivation
platform

Turn AI responses into a visual performance

Your AI
One AI response
Synchronise in real time
Captivation
In the browser
Lip sync
Gestures
Fillers
  • A smiling woman with blonde hair and glasses holding a white coffee cup with a smiley face design inside a coffee shop.

    Avatar

    Give your existing AI chatbot a human-like visual presence.

  • Animated woman with blonde hair, glasses, and a black turtleneck in front of a blue background, likely a digital assistant or AI chatbot interface.

    Lip sync

    Speech and text is synchronised to the SuperAgent in real time.

  • A cartoon woman with blonde hair, glasses, and a black shirt, resting her chin on her hands, with a speech bubble and buttons reading 'For Vendors' and 'For Commerce'.

    Gestures

    Contextual gestures and expressions add meaning and personality to the response.

  • A animated woman with blonde hair, glasses, and a surprised expression covering her mouth with her hands, on a digital interface with buttons and text.

    Fillers

    Natural movements and responses keep the SuperAgent alive while the AI is thinking or preparing its answer.

Captivation Inside FAQs

  • Captivation Inside is an embeddable Avatar layer for existing chatbots and AI agents. Your AI generates the response as normal; Captivation adds the voice, lip sync, expressions and contextual gestures and delivers the experience through an Avatar in your front end.

  • Yes. Captivation Inside is designed to be AI and LLM agnostic. You keep your existing AI, prompts, data, agents, tools and workflows and use Captivation as the visual and voice interface.

  • We provide a JavaScript widget for your front end and a connection to the Captivation Platform through the Captivation Bridge. Your LLM output is passed to Captivation, which generates the required voice and Avatar instructions and delivers them to the browser in real time through a WebSocket connection. Full technical details are provided in the developer guide.

  • Captivation can be used by businesses, developers and AI vendors. Developers can use Captivation Inside to add a face and voice to an existing chatbot or AI agent, while businesses can deploy complete SuperAgent solutions for ecommerce, sales, support and other customer-facing applications.

  • The time from receiving the LLM output to the Avatar beginning to speak is typically 200 to 300 milliseconds when stored audio can be reused, and 600 to 800 milliseconds when new audio must be created by a text-to-speech service. Actual performance can vary with the wider integration, network conditions and third-party services.

  • Yes. You can choose from Captivation's growing range of Avatars and voices. Enterprise clients can also commission bespoke Avatars and voices. Captivation can support compatible ElevenLabs voices for clients using their own ElevenLabs account.

  • The technology can work with most voice-agent architectures, including services such as Vapi and LiveKit. Captivation's fully supported commercial offering for this use case is still being developed, so please contact the team to discuss a specific requirement.