Layercode

Build production-ready voice AI
agents with TypeScript and Next.js

Layercode is voice AI infrastructure for developers. We handle WebSockets, voice activity detection, and deployment. You focus on your agent's logic.

$ npx @layercode/cli init
Start Building Read the docs

$100 free credits. No credit card required.

$ npx @layercode/cli init
Initializing project configuration...
✓ Created pipeline configuration
✓ Generated API route template
✓ Installed dependencies
Your voice agent is ready. Run npm run dev to start.

"Voice AI has unique infrastructure demands that traditional cloud architectures aren't built for. By leveraging Cloudflare, the Layercode team delivers the most performant and low-latency voice AI platform that scales."

DK
Dane Knecht
CTO at Cloudflare

Voice AI demos are easy.

Production is hard.

You built a working prototype in a weekend. But when real users start talking to your agent, everything breaks:

The agent lags and users talk over it
Turn-taking feels robotic
It mispronounces your customer's brand
Calls fail and you have no idea why
Scaling means rewriting your entire infrastructure

The gap between "cool demo" and production-ready voice AI is months of work: WebSocket connection management, voice activity detection tuning, global edge deployment, session recording, and observability tooling.

Layercode closes that gap.

Not a visual builder. Not a framework. Just infrastructure.

More control than Vapi or Retell

Visual workflow builders work until your logic gets complex. Then you're fighting the platform instead of building your product. Layercode gives you a webhook. Write TypeScript. Ship.

Simpler than LiveKit or Pipecat

Open-source frameworks give you control, but you're signing up for months of WebRTC, TURN servers, and audio pipeline debugging. Layercode handles the infrastructure. You handle the intelligence.

More flexible than OpenAI Realtime

Realtime LLM APIs are black boxes. You can't swap models mid-conversation, control prompts dynamically, or use your own fine-tuned LLM. Layercode calls YOUR backend. You control everything.

Your backend. Our infrastructure.

Layercode handles real-time audio streaming. You handle the conversation.

User speaks

Your user talks into their browser, phone, or mobile app. Layercode captures the audio stream at the nearest edge location and runs speech-to-text in real-time.

Your backend responds

We send transcribed text to your webhook. You process it with any LLM: OpenAI, Anthropic, Google, open-source models, etc. Stream your response back using our SDK.

User hears response

Layercode converts your text to speech and streams audio back to the user. The entire round-trip happens in under a second.

You receive text, you send text. No audio processing, no WebSocket management, no VAD tuning.

OpenAI, Claude, Gemini, Llama, Mistral, etc. Use whatever model fits your use case.

Vercel, AWS, Railway, your own servers. Layercode connects to it via webhook.

See the code

Add voice to your Next.js app in under 50 lines

Layercode's Node.js SDK integrates with the tools you already use. Here's a complete voice agent backend using the Vercel AI SDK: 

import { createOpenAI } from "@ai-sdk/openai";
import { streamText } from "ai";
import { streamResponse, verifySignature } from "@layercode/node-server-sdk";

const openai = createOpenAI({ apiKey: process.env.OPENAI_API_KEY! });

export const POST = async (request: Request) => {
  const body = await request.json();

  return streamResponse(body, async ({ stream }) => {
    if (body.type === "message") {
      const { textStream } = streamText({
        model: openai("gpt-4o-mini"),
        system: "You are a helpful voice assistant.",
        messages: [{ role: "user", content: body.text }],
        onFinish: () => stream.end(),
      });
      
      await stream.ttsTextStream(textStream);
    }
  });
};

Works with the LLM libraries you already use:

Vercel AI SDK
LangChain
OpenAI
Anthropic
Google AI
LlamaIndex
Ollama
Groq
View the full quickstart guide

Everything you need to ship voice AI

Hot-swap voice providers

Avoid vendor lock-in: Switch between Deepgram, ElevenLabs, Cartesia and Rime with a single config change. Test different models, optimize for cost or quality.

Analytics & Observability

Replay any conversation. Inspect latency breakdowns, and view transcripts to debug production issues.

Session recording

Every call is recorded automatically. Download audio files, export transcripts, build training datasets. All stored securely.

Per-second billing

Pay only for active conversation time. Silence is always free. No minimum commitments.

Web, mobile, and phone

Connect users via browser, iOS, Android, or phone. Same backend, same pipeline, multiple channels.

Unified billing

One invoice for speech-to-text, text-to-speech, and infrastructure.

"Layercode made it very easy to build and prototype low-latency voice features for our text-based agents built with NextJS and React."

LJ
Lance Jones
AI Agent Developer

330+ edge locations. Sub-50ms audio processing.

The first voice AI infrastructure built for low-latency conversations at global scale.

Traditional voice AI platforms run on centralized cloud infrastructure. When your user is in Tokyo and your servers are in Virginia, latency kills the conversation. Pauses feel unnatural. Users talk over the agent. The experience falls apart. Layercode is built on Cloudflare's global edge network. We process audio at the location nearest to your user—not in a distant data center.

Low-latency by default

Users connect to the nearest edge location. Speech-to-text, voice activity detection, and audio streaming happen locally.

Instant, automatic scaling

No capacity planning. No provisioning. Every conversation runs in its own isolated environment.

No noisy neighbors

Platform traffic spikes don't affect your users. Each session runs in complete isolation.

Global without configuration

Deploy once, serve users everywhere. No multi-region setup, no latency-based routing rules.

330+ Edge locations worldwide
<50ms Audio processing at edge
Zero Cold starts
100% Session isolation

Enterprise-ready security

SOC 2 Type II
GDPR Compliant
Data encrypted in transit (TLS 1.3)
Data encrypted at rest (AES-256)

Layercode is built for production workloads with enterprise security requirements. Your data is encrypted in transit and at rest. Session recordings are stored securely in SOC 2 compliant infrastructure and accessible only to your team.

Talk to us →

Simple, predictable pricing

Per-second billing for active conversation time. Silence is free. STT, TTS, and infrastructure costs consolidated into one simple rate. Start with $100 in free credits—no credit card required.

View pricing details