Embed Legacy Digital Humans in your Application

UNITH Digital Humans are platform agnostic. The following guidelines enable Digital Humans to be seamlessly integrated into various digital applications, from mobile apps to web apps.

info

If your digital human uses streaming mode, please refer to our SDK documentation instead.

To check whether your digital human is in legacy or streaming mode, see this page.

If the digital human configuration states that videoStreaming=true then your digital human is in streaming mode.

You should be able to access your organisation and digital human using USER/ME/endpoint. If you don't have access to the Digital Human platform please refer to User

code

curl -X 'GET' \
  'https://platform-api.unith.ai/user/me' \
  -H 'accept: application/json' \
  -H 'Authorization: Bearer bearer'

Code Snippet

Include the following code snippet in your website or application. These are the basic elements required to place digital humans in your page, but some additional code could be needed.

code

<!-- JavasScript bundle -->
<script defer src="https://embedded.unith.ai/dist/bundle.js"></script>

<!-- CSS bundle -->
<link rel="stylesheet" href="https://embedded.unith.ai/dist/bundle.css" />

<!-- Fonts -->
<link rel="stylesheet"
  href="https://gpt-head-assets.unith.ai/fonts/stylesheet.css"
  media="print"
  onload="this.media='all'"
/>

<!-- Microsoft speech recognition bundle -->
<script src="https://embedded.unith.ai/microsoft-speech-recognition.js"></script>
<!-- Body containing element with id="talking-head" -->
<body>

  <div
      id="talking-head"
      data-api_base="https://chat-origin.api.unith.live"
      data-api_key="<your-api-key>"
      data-embed="true"
      data-org_id="<your-org-id>"
      data-head_id="<your-head-id>"
      data-theme="demo"
      data-vm_version="1"
      data-time_pressure="false"
      data-subs_enabled="true"
      data-mic_enabled="true"
      data-seamless_transitions="false"      
      data-widget_mode="false"
      data-username="Roman"
      data-tag="campaign-summer2026"
  ></div>
</body>

Configure your Digital Human

The following configuration options are available to provide added controls and customisation options. These can be added inside the talking-head element.

data-embed: if set to something other "1" or "true", you will be able to access the full UI (default is true)
data-org_id: your organisation's friendly name, as shown in the PublicURL (required)
data-head_id: the head name of your Digital Human (required)
data-theme: currently the only supported value is "demo". We will support a variety of themes in the future.
data-vm_version: currently the only supported value is "1". This parameter controls the event handling.
data-time_pressure: The "Time Pressure" feature limits user response time to 60 seconds, after which a random suggestion is automatically selected.
data-mic_enabled: If set to something other than "1" or "true", the UI will not use speech recognition. No mic permissions will be needed from the user (default is true)
data-subs_enabled: If set to something other than "1" or "true", the UI will not display subtitles. (default is true)
data-language(optional): overrides the language configuration in your Digital Human for UI elements (default is en-US). If you need a language we do not support, contact us.
data-widget_mode: Setting this parameter totrue enables widget mode, embedding the Digital Human as a circular icon in the bottom right corner of the web application. The default value is false. Please refer to the full documentation on widget mode here.
data-seamless_transitions: Enabling the "Seamless Transitions" parameter eliminates noticeable jumps between the Digital Human's idle and response states.
data-welcome_message="VALUE" : Setting this parameter to a string enables you to customize the welcome screen headline message.
data-welcome_button_text="VALUE" : Setting this parameter to a string enables you to control the button placeholder text of the welcome screen.

Please to this documentation for further details.

warning_amber

id="talking-head-id" is not optional and the digital human won't work if removed.

User Identification & Session Tagging

You can associate a username and a custom tag with each legacy digital human session. Both values appear as dedicated columns in your conversation logs, enabling user-level tracking and custom segmentation without parsing session IDs.

data-username="VALUE" : The Custom Username feature enables you to assign a unique identifier to each user session interacting with your Digital Human. This username is then recognized and potentially used by the Digital Human during the conversation.
data-tag="VALUE" : Setting this parameter to a string enables you to add additional custom data to your digital human.

For legacy embeds, these are passed as data- attributes on the talking head element — URL parameters are not supported in this integration method:

code

<talking-head
  id="talking-head"
  data-org-id="your-org-id"
  data-head-id="your-head-id"
  data-username="roman"
  data-tag="campaign-summer2026"
></talking-head>

info

For full details on filtering logs by username and tag, see Conversation Logs Retrieval

Styling

Leverage CSS as needed to customize embedded element. The below acts as a guideline:

code

body {
      display: flex;
      flex-direction: column;
      align-items: center;
      height: auto;
    }
    p {
      text-align: center;
    }
    body > p {
      padding: 0 30%;
    }
    #talking-head {
      width: 900px;
      height: 600px;
      resize: both;
      overflow: auto;
      box-shadow: 2px 2px 14px -2px rgba(0, 0, 0, 0.2);
      border-radius: 10px;
      background-color: antiquewhite;
    }

info

Below you can find the current sitemap structure of our assets.

code

├── assets
│   ├── avatar_character.png
│   ├── avatar_user.png
│   ├── beep_up.wav
│   ├── favicon-192x192.png
│   ├── favicon.ico
│   ├── head-bg-large.png
│   ├── head-bg-small.png
│   ├── maya-large.png
│   ├── maya.png
│   └── noise-small.png
├── dist
│   ├── bundle.css
│   └── bundle.js
├── lang
│   ├── ar-AE.json
│   ├── bg-BG.json
│   ├── bs-BA.json
│   ├── cs-CZ.json
│   ├── de-DE.json
│   ├── en-US.json
│   ├── es-ES.json
│   ├── fr-FR.json
│   ├── hu-HU.json
│   ├── id-ID.json
│   ├── it-IT.json
│   ├── ka-GE.json
│   ├── kk-KZ.json
│   ├── lt-LT.json
│   ├── lv-LV.json
│   ├── nl-NL.json
│   ├── pl-PL.json
│   ├── pt-PT.json
│   ├── ro-RO.json
│   ├── ru-RU.json
│   ├── sk-SK.json
│   ├── sr-RS.json
│   ├── th-TH.json
│   └── uk-UA.json
├── README.md
├── manifest.json
├── microsoft-speech-recognition.js
└── themes
    └── demo.json

scheduleLast updated Apr 16, 2026