What Is a PAL?

A PAL (Personified Application Layer) is what you build in Conversational Video Interface (CVI). Configure its Face to define how it looks, its Voice to define how it sounds, and its behavior, knowledge, objectives, guardrails, tools, and pipeline layers to shape how it thinks and acts. Face and Voice are separate primitives that you configure as part of the PAL. A Conversation is the live session that brings the configured PAL to a participant over WebRTC.

Piece	What it is	What you configure
PAL	What you build in CVI	Behavior, knowledge, objectives, guardrails, tools, skills, and pipeline layers
Face	How the PAL looks	A stock Face or a trained Face
Voice	How the PAL sounds	A Tavus Voice on the PAL’s TTS layer or the Face’s default
Conversation	One live call	Which PAL joins, plus session options

Why the name matters

Docs and the API use PAL for the product you configure, Face for how it looks, and Voice for how it sounds. Older docs and some endpoints still say persona and replica. Same ideas: the PAL owns the configuration, while Face and Voice are separate primitives it uses. Prefer PAL and Face in new work. Legacy names still work on existing endpoints and request fields (/v2/personas, /v2/replicas, persona_id, replica_id, and related aliases) for backward compatibility.

Where to go next

PAL Overview - fields, layers, and customization options
What Is CVI? - how PAL, Face, and Conversation fit in the full pipeline
CVI Quickstart - create a PAL and start a conversation

Getting Started

Guides

Build

Deploy

Debug

Resources

What Is a PAL?

Why the name matters

Where to go next

​Why the name matters

​Where to go next

Why the name matters

Where to go next