#Can I integrate the conversational AI and voice into my own website?

1 messages · Page 1 of 1 (latest)

sleek violetBOT
#
AI-generated Answer for thejaysmith__

Yes. You have two supported ways to add ElevenLabs Conversational AI with voice to your own website:

  • Drop-in widget (fastest): Paste a small embed into your site to render a voice-enabled assistant. The docs provide the tag and loader script, plus guidance on allowing specific domains and other widget attributes. Note: the widget currently requires a public agent (authentication disabled) in your agent’s Advanced settings. (elevenlabs.io)

  • Build your own UI with SDKs or the realtime API: Use the React SDK or the framework-agnostic Client SDK to start sessions to your agent and stream audio; connect over WebRTC (recommended) or WebSocket, including options for signed URLs/tokens when using private agents. Direct WebSocket integration is also documented if you don’t want an SDK. (elevenlabs.io)

Helpful details:

  • Web widget/SDK embed example and customization options (appearance, text, multimodality, domain allowlist). (elevenlabs.io)
  • WebRTC support improves echo cancellation and background noise removal in browsers; enable by choosing connectionType "webrtc" in SDKs. (elevenlabs.io)
  • If you just want narrated articles (not a two‑way assistant), you can also use Audio Native, an embeddable player that voices page content. (help.elevenlabs.io)

If you tell me your stack (plain HTML, React/Next.js, Webflow/WordPress, etc.), I can point you to the exact guide. (elevenlabs.io)

Sources:
Widget customization | ElevenLabs Documentation
React SDK | ElevenLabs Documentation
Widget customization | ElevenLabs Documentation
ElevenLabs Conversational AI now supports WebRTC | ElevenLabs
What is Audio Native? – ElevenLabs
Conversational AI in WordPress | ElevenLabs Documentation

iron dew
#

I mean the ElevenLabs one. I’m using an LLM to create the website

sleek violetBOT
#
AI-generated Answer for thejaysmith__

Yes. You can add ElevenLabs Conversational AI (with voice) to your own site in a few supported ways:

  • No‑code web widget: Embed the <elevenlabs-convai> element and the widget script on your page. Note: as of today, widgets require a public agent with authentication disabled; use the domain allowlist for access control. See widget embed and options. Also see the quickstart for creating/configuring an agent. (elevenlabs.io)

  • JavaScript/React SDKs: Build a fully custom UI that streams the mic to your agent over a real‑time connection. Use @elevenlabs/client (vanilla JS) or @elevenlabs/react (React hook) to start/end sessions and handle events. (elevenlabs.io)

  • Private agents (auth): If you don’t want a public agent, generate a short‑lived Signed URL from your backend and pass it to the client when starting a session (example backend + client shown in the Vite/JS quickstart). (elevenlabs.io)

  • Bring your own LLM: You can point an agent at your own OpenAI key or a custom OpenAI‑compatible “chat completions” server. The docs include a reference server and setup steps. (elevenlabs.io)

  • Add tools and actions: Give the agent capabilities via Server Tools (function calls/Webhooks to your APIs) and Client Tools (perform actions in the user’s browser like navigation). (elevenlabs.io, help.elevenlabs.io)

  • Voice configuration: Choose voices, control speed/pronunciation, or enable multi‑voice behavior in agent settings. (elevenlabs.io)

  • Browser experience: Conversational AI now supports WebRTC in modern browsers for better echo cancellation/noise handling and reliability. (elevenlabs.io)

Platform guides are available if you’re using a site builder (e.g., WordPress/Webflow/Wix/Squarespace). (elevenlabs.io)

If you tell me your stack (static HTML, React/Next.js, WordPress, etc.), I can point you to the exact snippet or SDK example. (elevenlabs.io)

iron dew
#

Well I’m going to be using Figma Make AI to build the website to an API would be best