Skip to main content
Anam AI Avatars Anam builds real-time AI avatars: photorealistic digital humans that talk, listen, and perform actions based on your conversations. A persona is the combination of a face, a voice, an LLM, and a system prompt. You integrate a persona with your webpage and it holds live conversations with your users.

​
Who builds with Anam

Teams use Anam to build interactive agents from scratch, or to put a face on agents they already have. Common cases include customer support, sales and lead qualification, language tutoring, skill training, and medical front-desk assistance. In each of these cases, talking with an Anam avatar feels closer to a face-to-face conversation than text or voice alone.

​
How it works

Every live conversation with a persona runs through a four-stage pipeline:
  1. Speech-to-text (STT) listens to the user
  2. An LLM decides what to say back
  3. Text-to-speech (TTS) turns the reply into audio
  4. Face generation turns the audio into a live video of the persona speaking
The default setup, which we call Turnkey, runs the whole pipeline for you. You can also bring your own LLM, your own STT, your own TTS, or hand us pre-generated audio and have us run face generation only. Whichever setup matches your stack, the persona and the avatar stream stay the same.

​
Three ways to start

Learn

Read how personas work. Avatars, voices, LLMs, and how they fit together.

Embed

Add an avatar to a site you already have. Widget, Player, or platforms like Framer and Shopify.

Build

Use the JavaScript or Python SDK to build your own. The pipeline and the UI are yours.
Last modified on April 14, 2026