Reference

clientConfiguration fields (per-session agent config)

These flags shape runtime behavior:

Field	Value	Meaning
`languageCode`	`"en"`	conversation language
`interruptionsEnabled`	`true`	barge-in allowed (user can talk over the avatar)
`isTapToTalk`	`false`	push-to-talk vs open-mic. A fixed, per-agent config choice, not a live per-session toggle. Exposed read-only via `KalturaAvatarSession#capabilities.tapToTalk`, and gates `startTapToTalk()`/`endTapToTalk()` client-side (see the `tapToTalkStart`/`tapToTalkEnd` row in the events catalog for why mixing modes is unsafe)
`showTranscription`	`false`	surface live captions in UI
`isWebSearchEnabled`	`false`	gates real web-search tools (→ `tool`/`tool_response` deltas). *When `false`, the agent can still say* it will "look that up" but no `tool` segment fires** — the search doesn't happen.
`isScreenShareEnabled` / `isCameraAnalysisEnabled`	`false`	screen-share / camera-vision features
`audioMode` / `phoneMode`	`false`	audio-only / telephony modes
`pauseConversationEnabled`	`false`	can pause the conversation
`shouldAggregateCurrentTurn`	`false`	turn-aggregation behavior
`forwardLoopMode` / `imaginativeAiMode`	`false`	server-side conversation modes
`initialHtml`, `youtubeUrl`, `visualPhotos[]`, `visualVideos[]`	empty	initial GenUI content the agent ships with
`agentPersonaName`, `userName`	`null`	display names
`configuration`, `nluFeatures`	`{}`	extension buckets

Structured experiences (`force_experience` + `unisphere-tool`)

Scope: the structured-experience behavior below applies to the HTTP /assistant/converse path (headless/text integrations). The avatar runtime does not use it. A live avatar session always runs as force_experience: 'avatar_only' and model_type: 'fast' regardless of what's configured, so it never emits flashcards/summarization widgets. Use the HTTP converse path (or a custom client) to drive structured experiences.

force_experience on converse (e.g. "flashcards") is a hint, not a guarantee. The brain decides which structured widget(s) to emit based on the prompt and the intellect (the agent's brain configuration — its prompts, tools, and capabilities). Each comes back as unisphere-tool segments. The first carries metadata:{ widgetName, runtimeName }, then the content streams as a YAML-ish block (e.g. title: / questions:). For example:

force_experience:"flashcards" + a teachable prompt ("Teach me about video codecs") → both flashcards-tool and followups-tool runtimes in one turn (widgetName:"unisphere.widget.genie").
The same force_experience + a vague prompt ("show me something interesting") → only followups-tool.
So: render whatever runtimeName arrives; don't assume force_experience maps 1:1 to a widget. capabilities.generate_followup_questions:"on" independently yields the followups-tool.

Doc	Covers
Wire Protocol · Events Catalog	The `join`/`clientConfiguration` events these fields ride on
Wire Protocol	Back to the index