Real-time voice and multimodal AI web SDK
Top 97.2% on sourcepulse
This repository provides the official web client SDKs for Pipecat, an open-source Python framework for building real-time voice and multimodal AI applications. It targets frontend developers and researchers looking to integrate voice-based AI interactions into web applications, offering robust handling of media streams, bot configuration, and session management.
How It Works
The SDK is structured as a monorepo with two main packages: client-js
for a core JavaScript/TypeScript client and client-react
for React-specific components and hooks. It abstracts away the complexities of device and media stream management, bot communication, and session state. A key design principle is the separation of concerns, requiring an external "transport" package (e.g., @pipecat-ai/daily-transport
for WebRTC via Daily) to handle the actual data transmission, allowing flexibility in choosing communication protocols.
Quick Start & Requirements
npm install @pipecat-ai/client-js
npm install @pipecat-ai/client-react
npm install @pipecat-ai/daily-transport
Highlighted Details
Maintenance & Community
Licensing & Compatibility
Limitations & Caveats
The SDK does not include any transport capabilities out-of-the-box, necessitating the installation and configuration of separate transport packages. Server-side endpoints are recommended for secure handling of API keys and authentication.
23 hours ago
Inactive