Developers · SDK
Work in progressLiveSpeech SDK.
The SDK is for teams that write the code that talks: your own voice app, a web or mobile agent, or a phone line you already run. It opens a live speech session over WebSocket, streams audio both ways, and makes interruption control, tool calling, and conversation memory first-class APIs. Running phone lines for your customers instead? That is the Partner API.
Quickstart
Three lines of code to a live voice session.
Install the SDK
TypeScript via npm, Rust via crates.io.
Claim a number
A free 1-click phone number in Korea, the US, or Canada — from the console.
FAQ
Free numbers, languages, and integration questions, answered.
Install
# TypeScript
npm install @drawdream/livespeech
# Rust
cargo add livespeech-sdkPackage pages: @drawdream/livespeech on npm · livespeech-sdk on crates.io.
Quickstart — three lines
import { LiveSpeechClient } from '@drawdream/livespeech'
const client = new LiveSpeechClient({ apiKey: process.env.LIVESPEECH_KEY })
await client.startSession({ prePrompt: 'You are a scheduling agent.' })Sessions stream over WebSocket. The package READMEs cover interruption control, tool registration, and conversation memory.
Start freePutting an AI agent on a phone line for each of your customers? Go to the Partner API docs