Voiceblox is a visual builder for voice AI agents. You drag nodes onto a canvas, connect them, and get a fully working voice agent with no boilerplate required. It's backed by LiveKit's real-time infrastructure.

Do I need to write code to build a voice agent?

No. Voiceblox is a drag-and-drop builder. You describe your agent in plain language and the AI generates the conversation flow for you. You can also build manually on the canvas without writing any code.

What AI providers does Voiceblox support?

Voiceblox supports 14+ LLM providers (OpenAI, Gemini, DeepSeek, Groq, etc.), 8 TTS providers (ElevenLabs, Cartesia, Deepgram, etc.), and 7 STT providers (Deepgram, AssemblyAI, ElevenLabs, etc.). More providers are added regularly.

How do I deploy my voice agent?

Connect your LiveKit Cloud account in settings, then click Deploy in the builder. Voiceblox creates a LiveKit room and dispatches the agent worker automatically. You can also generate a public share link so anyone can talk to your agent without any setup.

What is LiveKit and why does Voiceblox use it?

LiveKit is a real-time audio/video infrastructure. Voiceblox uses it as the agent runtime: it handles the low-latency audio pipeline between the user and the AI, including STT, LLM, and TTS processing.

Can I build AI phone agents with Voiceblox?

Yes. Voiceblox supports SIP transfer nodes for inbound and outbound phone calls via LiveKit SIP. You can design branching call flows, route callers to specialists, and handle triage logic visually.

Can my agent collect information before the call?

Yes. Add input variables to your agent and Voiceblox shows a short form before the conversation starts. Callers fill in details, or upload a PDF like a CV, and the agent uses them during the call.

What do I get after each call?

Define the fields you care about and Voiceblox extracts them from every conversation as structured data, then delivers it via webhook, Zapier, or Make. The full transcript and form inputs can be included too.

The Lovable of Voice Agents

One voice agent runs the whole workflow

Collect the details you need before the call, run the conversation, and get a structured report after, all in one place. No code.

Talk to an expert

One agent, the whole workflow

From the first form field to the final report, Voiceblox runs every step.

Collect

Ask for what you need before the call starts: names, context, even a PDF or CV upload. Your agent reads it all going in.

Converse

A natural voice conversation handles the interaction, following the flow you designed on the canvas.

Report

Every call ends with structured data delivered where you work, via webhook, Zapier, or Make.

The tightest feedback loop

Spin it up. Call it. Refine until it's right.

Generate from a prompt

Describe the voice agent you want, like 'Build a pitch practice bot', and the AI creates the full conversation flow instantly.

Edit visually on the canvas

Drag, drop, and connect nodes to reshape the conversation flow exactly how you want it.

Edit it by prompting

Describe what you want to change in plain language and watch the agent update in real time.

Test it immediately

Talk to your agent right inside the builder - no deployment needed.

Improve a response you didn't like

When the agent says something off, click on that message in the conversation panel to correct it.

Deploy directly from the app

When your agent is ready, deploy it with a single click. No extra tools or setup required.

Share it with anyone

Generate a share link and send it to anyone. They can test your agent directly on voiceblox.ai, no app install needed.

Everything you need

A complete toolkit for building voice AI agents.

Large Language Models

18 providers live including OpenAI, Anthropic, and Mistral AI. 3 more coming soon.

OpenAI

Gemini

Moonshot

DeepSeek

Groq

Together AI

Azure

Cerebras

Fireworks

Ollama

OVHCloud

Perplexity

Telnyx

xAI

Anthropic

Mistral AI

OpenRouter

Amazon Bedrock

Baseten

Letta

LangChain

Speech to Text

11 providers live including Deepgram, Google Cloud, and ElevenLabs. 12 more coming soon.

Deepgram

ElevenLabs

Cartesia

AssemblyAI

Groq

OpenAI

OVHCloud

Google Cloud

Mistral AI

AWS

Azure Speech

Whisper

Azure OpenAI

Baseten

Clova

fal

Gladia

Gradium

Nvidia

Sarvam

Simplismart

Soniox

Speechmatics

Spitch

Text to Speech

12 providers live including ElevenLabs, Google Cloud, and Groq. 16 more coming soon.

ElevenLabs

Deepgram

Cartesia

Inworld

Rime

OpenAI

Neuphonic

Resemble AI

Google Cloud

AWS

Groq

Azure Speech

PlayHT

Gemini

Azure OpenAI

Nvidia

Baseten

Gradium

Hume

LMNT

MiniMax

Sarvam

Simplismart

Smallest AI

Speechify

Spitch

AsyncAI

Camb.ai

Avatar

10 providers live including LemonSlice, Tavus, Simli, and more. 2 more coming soon.

Anam

Avatario

AvatarTalk

Hedra

Keyframe

LemonSlice

LiveAvatar

Simli

Tavus

TruGen

Beyond Presence

bitHuman

Conversation Flow

Start, Timer, Open Talk, Burst, If/Else, Categorize, and End are live. Transfer coming soon.

Start

Timer

Open Talk

Burst

If/Else

Categorize

End

Transfer

Post-processing

Webhook, Structured Output, Zapier, and Make are live. Extract structured data and deliver it after each session.

Webhook

Structured Output

Zapier

Make

Summarization

CRM

Helpdesk

Tools

MCP, Exa, and SIP Transfer are live. Knowledge base, calendar, CRM, and more coming soon.

MCP

Exa

Pinecone

SIP Transfer

Human Handoff

Knowledge Base

Calendar

CRM

SMS

Prefer to have it built for you?

Skip the build. Tell me what you need and I will design, build, and deploy your voice agent.

Book a call

Frequently asked questions

Everything you need to know about building voice agents with Voiceblox.

Voiceblox

The Lovable of Voice Agents

One voice agent runs the whole workflow

Collect the details you need before the call, run the conversation, and get a structured report after, all in one place. No code.

Talk to an expert