A collection of example applications built with Pipecat, an open-source framework for building voice and multimodal AI applications.
Learning examples are in the main Pipecat repo, intermediate and advanced examples are here.
Start with the quickstart example in the main Pipecat repo to get your first bot running in 5 minutes.
Then continue learning with these starter examples, located in the Pipecat repo:
- client-server-web example: Learn client/server architecture with web clients
- phone-bot-twilio: Learn how to connect your bot to a phone number using Twilio
Once you understand the basics, check out the examples below.
Most examples require:
- Python 3.10 or newer
- API keys for AI services (OpenAI, Deepgram, Cartesia, etc.)
- Additional service-specific requirements (see individual example READMEs)
Ready to explore more? These are two of the most useful examples for common use cases:
- simple-chatbot - Client/server examples with React, JavaScript, Swift, Kotlin, and React Native
- twilio-chatbot - Production-ready phone bot with Twilio integration
- phone-chatbot - Daily PSTN and SIP dial-in and dial-out examples
- twilio-chatbot - Production-ready phone bot with Twilio integration
- telnyx-chatbot - Production-ready phone bot with Telnyx integration
- plivo-chatbot - Production-ready phone bot with Plivo integration
- simple-chatbot - Client/server examples with React, JavaScript, Swift, Kotlin, and React Native
- websocket - WebSocket-based real-time communication
- instant-voice - Enable instant voice communication as soon as a user connects
- p2p-webrtc - Simple peer-to-peer WebRTC voice bot
- word-wrangler-gemini-live - Web & phone based voice AI word games using Gemini Live
- aws-strands - Use AWS Strands for multi-step tasks
- moondream-chatbot - Vision + voice multimodal AI bot
- storytelling-chatbot - Interactive storytelling experiences
- news-chatbot - AI news reader and discussion bot
- translation-chatbot - Real-time language translation
- daily-multi-translation - Multi-language conference calls
- patient-intake - Healthcare intake automation via phone
- studypal - AI-powered study companion
- chatbot-audio-recording - Voice interactions with recording capabilities
- fal-smart-turn - Smart conversation turn-taking
- daily-custom-tracks - Custom audio/video track handling
- local-input-select-stt - Local audio input selection
- bot-ready-signalling - Bot connect management
- deployment - Production deployment examples using Pipecat Cloud, Fly.io, Modal, Cerebrium
- sentry-metrics - Use Sentry to collect Time-to-First-Byte and processing metrics
- open-telemetry - Observability and tracing examples using Langfuse and Jaeger
- freeze-test - Pipeline freezing and state management testing
- Documentation: docs.pipecat.ai
- Discord Community: discord.gg/pipecat
- Main Repository: github.com/pipecat-ai/pipecat
- Issues: Report bugs or request examples in the main repo issues