Pipecat
Pipecat is an open-source Python framework for building real-time voice and multimodal conversational agents. It solves the complexity of integrating speech recognition, text-to-speech, LLMs, and transports into composable pipelines, enabling developers to create voice assistants, multi-agent systems, and more. It is designed for developers building AI-powered voice and multimodal applications.
✨ Key features
- Voice-first: integrates STT, TTS, and conversation handling
- Pluggable: supports many AI services and tools
- Composable pipelines: build complex behavior from modular components
- Multi-agent ready: compose agents with handoff, fan-out, sidecars
- Real-time: ultra-low latency with WebSockets or WebRTC
- Client SDKs for JavaScript, React, Swift, Kotlin, C++, ESP32
🎯 Use cases
- Build voice assistants with natural streaming conversations
- Create multi-agent systems with specialist handoffs and parallel fan-out
- Develop AI companions like coaches or meeting assistants
- Implement multimodal interfaces combining voice, video, and images
- Design interactive storytelling and business agents
📦 Installation
🧰 Requirements: Python environment; optional API keys for AI services; client SDKs for platform apps.
pip install pipecat-ai
For CLI: uv tool install "pipecat-ai[cli]"
🚀 Usage
from pipecat.pipeline import Pipeline
from pipecat.transports import ...
# Example from README not provided; see quickstart guide.
❓ FAQ
What is Pipecat?
Pipecat is an open-source Python framework for building real-time voice and multimodal conversational agents.
How do I install Pipecat?
Install with pip install pipecat-ai or use the CLI with uv tool install "pipecat-ai[cli]".
What can I build with Pipecat?
You can build voice assistants, multi-agent systems, AI companions, multimodal interfaces, and more.
Does Pipecat support multiple AI services?
Yes, it supports many STT, LLM, and TTS services from providers like OpenAI, Google, and Deepgram.
📊 Repository
🤖 Overview, features, install steps and FAQ were generated from the project's README on Sep 4, 2026. Always check the original source before running commands.