Skip to main content
XUNA AI Conversational AI is a complete platform for building voice-first AI agents that accomplish tasks through natural dialogue. From a quick customer support bot to complex, multi-step workflows, you get everything you need — expressive speech models, developer SDKs, deployment integrations, and analytics — in one place.

Quick Start

Build and test your first voice agent in under 5 minutes.

Configure Your Agent

Set up prompts, voices, knowledge bases, and tools.

Deploy Everywhere

Embed on the web, integrate with phone systems, or use native SDKs.

API Reference

Explore the REST API and WebSocket protocol.

​
Platform capabilities

XUNA AI Conversational AI coordinates four core components to deliver natural, low-latency voice conversations:
  1. Speech recognition — A fine-tuned ASR model transcribes user speech in real time.
  2. Language model — Your choice of LLM (GPT, Gemini, Claude, or a custom model) generates responses.
  3. Voice synthesis — A low-latency TTS engine renders responses across 5,000+ voices and 70+ languages.
  4. Turn-taking — A proprietary model manages conversation timing, interruptions, and silence handling.

​
What you can build

Customer Support

Automate support with agents that answer questions, resolve tickets, and escalate intelligently.

Sales & Outbound

Run outbound campaigns at scale with batch calling and personalized conversation flows.

Voice Interfaces

Add voice interaction to any app using the React, iOS, or Android SDKs.

​
How to get started

1

Sign up

Create a free account at xuna.ai.
2

Create an agent

Use the dashboard, CLI, or API to configure your first agent. Follow the Quick Start guide for a step-by-step walkthrough.
3

Deploy

Embed a widget on your website, connect a phone number, or integrate via SDK. See Deploy for all options.
4

Monitor and improve

Review conversation transcripts, run A/B experiments, and analyze outcomes in the analytics dashboard.