# Pipecat > Pipecat is an open source ecosystem for building voice and multimodal AI > agents. Build with the Python framework, connect users with the client SDKs, > structure conversations with Pipecat Flows, and deploy to production on > Pipecat Cloud. ## About Pipecat Pipecat is an open source framework and hosted platform for real-time voice and multimodal AI agents. Its primary components: - **Pipecat framework**: An open source Python framework that orchestrates AI services (STT, LLM, TTS, and more) into real-time pipelines. Start with the [quickstart](https://docs.pipecat.ai/pipecat/get-started/quickstart.md). - **Pipecat client SDKs**: JavaScript, React, React Native, iOS, Android, and C++ SDKs that connect users to agents over WebRTC or WebSockets. See the [client introduction](https://docs.pipecat.ai/client/introduction.md). - **Pipecat Flows**: A framework for structured conversations - define conversation paths as nodes with functions and actions. See the [Flows introduction](https://docs.pipecat.ai/pipecat-flows/introduction.md). - **Pipecat Cloud**: Managed infrastructure for deploying and scaling agents, run by the Pipecat team. See the [Cloud introduction](https://docs.pipecat.ai/pipecat-cloud/introduction.md). - **Pipecat CLI**: Scaffold projects, run evals, and deploy from the terminal. See the [CLI overview](https://docs.pipecat.ai/api-reference/cli/overview.md). Server API reference lives under `api-reference/server/`: services (STT, TTS, LLM, transports, serializers), pipelines, frames, workers, and utilities. ## Overview ### Overview - [Pipecat Documentation](https://docs.pipecat.ai/overview/introduction.md): Documentation for the Pipecat ecosystem: the open source framework, client SDKs, Pipecat Flows, and Pipecat Cloud hosting. - [Pipecat Open Source Framework](https://docs.pipecat.ai/overview/pipecat.md): Pipecat is an open source Python framework for voice and multimodal AI agents, orchestrating AI services and transports. - [Pipecat Client SDKs](https://docs.pipecat.ai/overview/clients.md): The Pipecat client SDK family connects users to your agents from web and mobile apps, handling the real-time media layer. - [Pipecat Flows](https://docs.pipecat.ai/overview/flows.md): Pipecat Flows adds structured conversations to your voice agents: define conversation paths as nodes with functions and actions. - [Pipecat Cloud](https://docs.pipecat.ai/overview/cloud.md): Pipecat Cloud is the managed hosting platform for deploying and scaling Pipecat agents, from the team behind the framework. - [Enterprise Support](https://docs.pipecat.ai/enterprise-support/index.md): Enterprise support for Pipecat and Pipecat Cloud from Daily: architecture guidance, production readiness, and compliance. ## Pipecat ### Get Started - [Introduction to Pipecat](https://docs.pipecat.ai/pipecat/get-started/introduction.md): Start here with the Pipecat framework: what it is, how pipelines orchestrate AI services, and where to go next. - [Pipecat Quickstart](https://docs.pipecat.ai/pipecat/get-started/quickstart.md): Build and run your first Pipecat voice AI bot in under 5 minutes: a simple conversational agent you can talk to in a browser. - [Build Your Next Bot](https://docs.pipecat.ai/pipecat/get-started/build-your-next-bot.md): Scaffold a new Pipecat project and build it with a coding agent or by hand - [Continue Learning](https://docs.pipecat.ai/pipecat/get-started/next-steps.md): Where to go after the Pipecat quickstart: learning paths, examples, and guides for building production voice agents. ### Migration - [Migrating to Pipecat 1.0](https://docs.pipecat.ai/pipecat/migration/migration-1.0.md): Upgrade a Pipecat application from 0.0.x to 1.0: removed deprecated APIs, their replacements, and the migration path. ### Learning Pipecat - [Overview of Pipecat](https://docs.pipecat.ai/pipecat/learn/overview.md): Learn the foundational concepts of Pipecat's architecture for building voice AI agents - [Your First Agent](https://docs.pipecat.ai/pipecat/learn/your-first-agent.md): Build and run a single agent: a PipelineWorker, the WorkerRunner, and the pipeline lifecycle. - [Session Initialization](https://docs.pipecat.ai/pipecat/learn/session-initialization.md): Learn how to set up connections between users and your Pipecat voice AI bot - [Pipeline & Frame Processing](https://docs.pipecat.ai/pipecat/learn/pipeline.md): Learn how Pipecat's pipeline architecture orchestrates frame processing for voice AI applications - [Transports](https://docs.pipecat.ai/pipecat/learn/transports.md): Learn about the different ways users can connect to your Pipecat voice AI bot - [Speech Input & Turn Detection](https://docs.pipecat.ai/pipecat/learn/speech-input.md): Learn how Pipecat detects user turns using VAD, transcriptions, and turn detection models - [Speech to Text](https://docs.pipecat.ai/pipecat/learn/speech-to-text.md): Learn how to configure speech recognition to convert user audio into text in your Pipecat pipeline - [Context Management](https://docs.pipecat.ai/pipecat/learn/context-management.md): Work with Pipecat's LLM context and context aggregators: how conversation history is built and shared in a pipeline. - [LLM Inference](https://docs.pipecat.ai/pipecat/learn/llm.md): Learn how to configure language models to generate intelligent responses in your voice AI pipeline - [Function Calling](https://docs.pipecat.ai/pipecat/learn/function-calling.md): Enable LLMs to interact with external services and APIs in your voice AI pipeline - [Text to Speech](https://docs.pipecat.ai/pipecat/learn/text-to-speech.md): Learn how to configure speech synthesis to convert text into natural-sounding audio in your voice AI pipeline - [Pipeline Termination](https://docs.pipecat.ai/pipecat/learn/pipeline-termination.md): Learn how to properly terminate Pipecat pipelines for clean shutdown and resource management - [Multiple LLM Agents](https://docs.pipecat.ai/pipecat/learn/multiple-llm-agents.md): Run separate agents that each own their own LLM, tools, and conversation context. - [Controlling the UI](https://docs.pipecat.ai/pipecat/learn/ui-worker.md): Bridge a voice agent and a client GUI with a UIWorker over a two-way RTVI interface. - [Agent Handoff](https://docs.pipecat.ai/pipecat/learn/agent-handoff.md): Transfer control between agents with activation, deactivation, and handoff. - [Job Coordination](https://docs.pipecat.ai/pipecat/learn/job-coordination.md): Dispatch work from one Pipecat agent to others and collect results with job coordination in the multi-agent framework. - [Distributed Agents](https://docs.pipecat.ai/pipecat/learn/distributed-agents.md): Run Pipecat agents across processes and machines connected to the same bus for distributed multi-agent deployments. - [Proxy Agents](https://docs.pipecat.ai/pipecat/learn/proxy-agents.md): Connect Pipecat agents running on different buses with proxy agents, extending multi-agent pipelines across processes. - [What's Next](https://docs.pipecat.ai/pipecat/learn/whats-next.md): Continue your Pipecat journey with advanced features, examples, and production deployment ### Fundamentals - [Service Settings](https://docs.pipecat.ai/pipecat/fundamentals/service-settings.md): Learn how to configure and update AI service settings at initialization and runtime. - [Interruptions](https://docs.pipecat.ai/pipecat/fundamentals/interruptions.md): How Pipecat stops the bot when the user speaks, what happens to in-flight LLM and TTS output, and how to control it. - [User Input Muting](https://docs.pipecat.ai/pipecat/fundamentals/user-input-muting.md): Learn how to control when user speech is processed in your conversational bot - [Detecting Idle Users](https://docs.pipecat.ai/pipecat/fundamentals/detecting-user-idle.md): Learn how to detect and respond when users are inactive in conversations - [STT Latency Tuning](https://docs.pipecat.ai/pipecat/fundamentals/stt-latency-tuning.md): Measure and tune STT latency in Pipecat to improve turn detection timing and end-of-turn responsiveness. - [Context Summarization](https://docs.pipecat.ai/pipecat/fundamentals/context-summarization.md): Automatically compress older conversation history in long-running Pipecat conversations to manage LLM token usage and cost. - [Saving Conversation Transcripts](https://docs.pipecat.ai/pipecat/fundamentals/saving-transcripts.md): Learn how to collect and save conversation transcripts between users and your bot - [Recording Conversation Audio](https://docs.pipecat.ai/pipecat/fundamentals/recording-audio.md): Learn how to record and save audio from conversations between users and your bot - [Metrics](https://docs.pipecat.ai/pipecat/fundamentals/metrics.md): Monitor Pipecat performance metrics: TTFB, processing time, and LLM and TTS usage across pipeline services. - [Voicemail Detection](https://docs.pipecat.ai/pipecat/fundamentals/voicemail.md): Automatically classify outbound calls as conversation or voicemail and respond appropriately - [IVR Navigation](https://docs.pipecat.ai/pipecat/fundamentals/ivr.md): Automatically navigate phone system menus using AI-powered decision making - [Custom FrameProcessor](https://docs.pipecat.ai/pipecat/fundamentals/custom-frame-processor.md): Write a custom Pipecat FrameProcessor: handle frames, push new ones downstream, and slot into a pipeline. #### Multi-Agent Architecture - [The Worker Bus](https://docs.pipecat.ai/pipecat/fundamentals/agent-bus.md): How Pipecat workers communicate through the shared message bus: publish, subscribe, and message flow between agents. - [Agent Registry and Discovery](https://docs.pipecat.ai/pipecat/fundamentals/agent-registry-and-discovery.md): How agents discover each other and get notified when other agents are ready. - [Understanding the Bus Bridge](https://docs.pipecat.ai/pipecat/fundamentals/understanding-the-bus-bridge.md): How BusBridgeProcessor routes frames between the transport pipeline and the agent bus. ### Evals - [Pipecat Evals](https://docs.pipecat.ai/pipecat/evals/overview.md): Behavioral testing for your agents: scripted conversations, semantic assertions, and an LLM judge. - [Evals Lifecycle](https://docs.pipecat.ai/pipecat/evals/lifecycle.md): How local Pipecat Evals fit with platform testing, monitoring, and team-scale evaluation. - [Evals Quickstart](https://docs.pipecat.ai/pipecat/evals/quickstart.md): Run your first Pipecat behavioral eval against an existing agent with pipecat eval and a simple scenario. - [Writing Scenarios](https://docs.pipecat.ai/pipecat/evals/scenarios.md): The scenario file format: configuration, turns, events, and assertions. - [Eval Suites](https://docs.pipecat.ai/pipecat/evals/suites.md): Spawn agents and run many scenarios concurrently from a single manifest. - [Using the Library](https://docs.pipecat.ai/pipecat/evals/library.md): Run, build, and orchestrate evals from Python with the pipecat.evals API. - [The Eval Loop](https://docs.pipecat.ai/pipecat/evals/the-eval-loop.md): Close the loop: Pipecat evals give an AI coding assistant a pass/fail signal it can read, so it writes agent code, runs evals, and iterates. #### Third-party Platforms - [Arize](https://docs.pipecat.ai/pipecat/evals/platforms/arize.md): Observability and online evaluation for Pipecat voice agents, powered by OpenInference auto-instrumentation and OpenTelemetry. - [Bluejay](https://docs.pipecat.ai/pipecat/evals/platforms/bluejay.md): Simulation, observability, and evaluation platform for voice AI agents with native Pipecat integration. - [Cekura](https://docs.pipecat.ai/pipecat/evals/platforms/cekura.md): Simulation and production monitoring platform for Pipecat agents with flexible connection methods and metric-driven QA. - [Coval](https://docs.pipecat.ai/pipecat/evals/platforms/coval.md): AI-native simulation and evaluation platform for voice agents, trusted by QA, Engineering, Operations, AI, and Executive teams. - [Roark Eval Platform](https://docs.pipecat.ai/pipecat/evals/platforms/roark.md): Test and monitor Pipecat voice agents with Roark: automated simulations with personas, plus observability, tracing, and metrics. ### Features - [Krisp VIVA](https://docs.pipecat.ai/pipecat/features/krisp-viva.md): Integrate Krisp VIVA into Pipecat: voice isolation, noise filtering, and turn detection from the Krisp VIVA SDK. - [WhatsApp Business Calling API](https://docs.pipecat.ai/pipecat/features/whatsapp.md): Receive WhatsApp voice calls in a Pipecat application using the WhatsApp Business Calling API and WhatsAppTransport. - [Building with Gemini Live](https://docs.pipecat.ai/pipecat/features/gemini-live.md): Create real-time voice AI agents using Google's Gemini Live API and Pipecat - [Building With OpenAI Audio Models and APIs](https://docs.pipecat.ai/pipecat/features/openai-audio-models-and-apis.md): Build voice agents with OpenAI audio models in Pipecat: STT, TTS, and the Realtime API, and when to use each. ### Telephony - [Pipecat Telephony Overview](https://docs.pipecat.ai/pipecat/telephony/overview.md): Give Pipecat bots phone capabilities: dial-in and dial-out over PSTN and SIP with Daily, Twilio, Telnyx, Plivo, and Exotel. - [Daily Phone Numbers](https://docs.pipecat.ai/pipecat/telephony/daily-phone-numbers.md): Purchase and manage phone numbers for Daily's PSTN services to give your Pipecat bots dial-in and dial-out. - [Daily PSTN](https://docs.pipecat.ai/pipecat/telephony/daily-pstn.md): Complete guide to Daily's PSTN capabilities including dial-in, dial-out, and call transfers - [Daily SIP](https://docs.pipecat.ai/pipecat/telephony/daily-sip.md): Use Daily as the SIP provider for dial-in and dial-out with any SIP-capable telephony carrier (Twilio, Telnyx, Plivo, etc.). - [Daily + Twilio SIP](https://docs.pipecat.ai/pipecat/telephony/twilio-daily-sip.md): Complete guide to using Daily's WebRTC transport with Twilio's SIP services for dial-in and dial-out - [Twilio WebSocket Integration](https://docs.pipecat.ai/pipecat/telephony/twilio-websockets.md): Complete guide to using Twilio Media Streams with Pipecat for dial-in and dial-out functionality - [Telnyx WebSocket Integration](https://docs.pipecat.ai/pipecat/telephony/telnyx-websockets.md): Complete guide to using Telnyx Media Streaming with Pipecat for dial-in and dial-out functionality - [Plivo WebSocket Integration](https://docs.pipecat.ai/pipecat/telephony/plivo-websockets.md): Complete guide to using Plivo Media Streaming with Pipecat for dial-in and dial-out functionality - [Exotel WebSocket Integration](https://docs.pipecat.ai/pipecat/telephony/exotel-websockets.md): Complete guide to using Exotel Voice Streaming with Pipecat for dial-in and dial-out functionality ### Deployment - [Pipecat Deployment Overview](https://docs.pipecat.ai/pipecat/deployment/overview.md): Run Pipecat bots locally and in production: process model, hosting patterns, and choosing a deployment approach. - [Running Bots Locally](https://docs.pipecat.ai/pipecat/deployment/running-bots-locally.md): The bot entry point, the built-in development runner, and the transports it supports. - [Running Bots in Production](https://docs.pipecat.ai/pipecat/deployment/running-bots-in-production.md): What plays the role of the development runner when you run bots in production, and how to think about the choices. #### Production Hosting Patterns - [VM per Session](https://docs.pipecat.ai/pipecat/deployment/patterns/vm-per-session.md): A dispatcher spawns a fresh cloud VM for each session via the provider's machines API. - [Warm Pool with Subprocess Workers](https://docs.pipecat.ai/pipecat/deployment/patterns/warm-pool-subprocess.md): Pre-allocate transport resources and bot subprocesses on a long-lived host; replenish on use. - [Managed Agent Runtime](https://docs.pipecat.ai/pipecat/deployment/patterns/managed-runtime.md): Hand the bot lifecycle off to a runtime that owns scaling, dispatch, and execution. - [Telephony in Production](https://docs.pipecat.ai/pipecat/deployment/telephony-in-production.md): Webhook-driven dispatch, carrier-specific gotchas, and how telephony differs from WebRTC. #### Hosting Platforms - [Fly.io](https://docs.pipecat.ai/pipecat/deployment/platforms/fly.md): Deploy Pipecat bots to Fly.io machines: app configuration, scaling, and running voice agents in production. - [Cerebrium](https://docs.pipecat.ai/pipecat/deployment/platforms/cerebrium.md): Deploy Pipecat voice agents to Cerebrium: configuration, deployment steps, and production setup. - [Modal](https://docs.pipecat.ai/pipecat/deployment/platforms/modal.md): Deploy Pipecat voice agents to Modal: configuration, deployment steps, and running bots in production. ### Examples & Recipes - [Pipecat Examples](https://docs.pipecat.ai/pipecat/examples/overview.md): Complete Pipecat example applications and quickstart demos: single-agent and multi-agent apps to accelerate development. - [Recipes](https://docs.pipecat.ai/pipecat/examples/recipes.md): Pipecat code recipes: focused snippets and techniques for common voice agent tasks and pipeline patterns. ## Pipecat Clients ### Get Started - [Introduction to Pipecat Clients](https://docs.pipecat.ai/client/introduction.md): Build real-time voice and multimodal apps with Pipecat client SDKs: microphone, playback, and transport handling built in. - [Pipecat Client Quickstart](https://docs.pipecat.ai/client/get-started/quickstart.md): Get a Pipecat client voice app running in minutes with the CLI, with React, JavaScript, and mobile SDK variants. ### Core Concepts - [The RTVI Standard](https://docs.pipecat.ai/client/rtvi-standard.md): The RTVI standard for real-time voice interaction: the open protocol connecting Pipecat clients and servers. - [Choosing a Transport](https://docs.pipecat.ai/client/concepts/choosing-a-transport.md): Pick the right transport for a Pipecat client app: Daily WebRTC, SmallWebRTC, and WebSocket options compared. - [Session Lifecycle](https://docs.pipecat.ai/client/concepts/session-lifecycle.md): How a Pipecat session starts, runs, and ends — and how to handle each phase. - [Events & Callbacks](https://docs.pipecat.ai/client/concepts/events-and-callbacks.md): How to respond to bot and session events in Pipecat client applications. - [Media Management](https://docs.pipecat.ai/client/concepts/media-management.md): Managing microphones, cameras, speakers, and media tracks in Pipecat client applications. ### Guides - [Building a Voice UI](https://docs.pipecat.ai/client/guides/building-a-voice-ui.md): Build a voice application UI from scratch with the Pipecat client SDKs: connection, audio, and bot interaction. - [Custom Messaging](https://docs.pipecat.ai/client/guides/custom-messaging.md): Send and receive arbitrary messages between your client and Pipecat bot. ### Voice UI Kit - [Voice UI Kit](https://docs.pipecat.ai/client/voice-ui-kit.md): Pre-built React components for building voice AI interfaces with Pipecat. ## Pipecat Flows ### Get Started - [Introduction to Pipecat Flows](https://docs.pipecat.ai/pipecat-flows/introduction.md): Learn Pipecat Flows, the framework for structured conversations: conversation paths, state management, and node transitions. - [Pipecat Flows Quickstart](https://docs.pipecat.ai/pipecat-flows/guides/quickstart.md): Build your first Pipecat Flows conversation: a two-node Hello World flow with a function call recording the user's answer. ### Guides - [Nodes & Messages](https://docs.pipecat.ai/pipecat-flows/guides/nodes-and-messages.md): Configure Pipecat Flows conversation nodes: role and task messages, and how nodes shape each conversation step. - [Functions](https://docs.pipecat.ai/pipecat-flows/guides/functions.md): Define function calls and handlers in Pipecat Flows to drive node transitions and collect structured data. - [Actions](https://docs.pipecat.ai/pipecat-flows/guides/actions.md): Execute custom functionality at specific points in your conversation flow. - [Context Strategies](https://docs.pipecat.ai/pipecat-flows/guides/context-strategies.md): Manage conversation context across Pipecat Flows node transitions: append, reset, and summarize strategies. - [State Management](https://docs.pipecat.ai/pipecat-flows/guides/state-management.md): Share data across Pipecat Flows nodes with FlowManager state and global functions for stateful conversations. ### Examples - [Pipecat Flows Examples](https://docs.pipecat.ai/pipecat-flows/examples.md): Pipecat Flows examples for common conversation patterns: structured dialogs, branching paths, and state management techniques. ### Migration - [Migrating to Pipecat Flows 1.0](https://docs.pipecat.ai/pipecat-flows/migration/migration-1.0.md): Upgrade a Pipecat Flows application from 0.0.x to 1.0: removed deprecated APIs, replacements, and the migration path. ## Pipecat Cloud ### Get Started - [Introduction to Pipecat Cloud](https://docs.pipecat.ai/pipecat-cloud/introduction.md): Pipecat Cloud is managed infrastructure for deploying and scaling Pipecat voice agents to production, run by the Pipecat team. ### Fundamentals - [Pipecat Cloud Accounts and Organizations](https://docs.pipecat.ai/pipecat-cloud/fundamentals/accounts-and-organizations.md): Manage your Pipecat Cloud account, invite teammates, and collaborate on agent deployments through shared organizations. - [Pipecat Cloud Agent Images](https://docs.pipecat.ai/pipecat-cloud/fundamentals/agent-images.md): Containerize a Pipecat agent for Pipecat Cloud: Dockerfile structure, base images, and image build requirements. - [Pipecat Cloud Deployments](https://docs.pipecat.ai/pipecat-cloud/fundamentals/deploy.md): Deploy agents with the pipecat cloud deploy CLI command: cloud image builds from your Dockerfile and deployment configuration. - [Starting Pipecat Cloud Sessions](https://docs.pipecat.ai/pipecat-cloud/fundamentals/active-sessions.md): Start on-demand Pipecat Cloud agent sessions: how sessions are created per request and scaled by your deployment configuration. - [Pipecat Cloud Secrets](https://docs.pipecat.ai/pipecat-cloud/fundamentals/secrets.md): Store API keys and sensitive configuration in Pipecat Cloud secret sets, and expose them to your agents as environment variables. - [Pipecat Cloud Scaling](https://docs.pipecat.ai/pipecat-cloud/fundamentals/scaling.md): Scale Pipecat Cloud agent deployments in production: instances, min-agents, and granular controls for compute and cost. - [Pipecat Cloud Health Checks](https://docs.pipecat.ai/pipecat-cloud/fundamentals/health-checks.md): Report agent readiness and liveness to Pipecat Cloud, and recycle agent instances on your own schedule with health endpoints. - [Pipecat Cloud Logging and Observability](https://docs.pipecat.ai/pipecat-cloud/fundamentals/logging.md): Obtain logs and metrics from Pipecat Cloud agents and sessions for debugging, monitoring, and production observability. - [Pipecat Cloud Troubleshooting](https://docs.pipecat.ai/pipecat-cloud/fundamentals/error-codes.md): Diagnose Pipecat Cloud deployment failures: common error codes, their causes, and fixes for agents that will not start or scale. ### Guides - [Pipecat Cloud Builds](https://docs.pipecat.ai/pipecat-cloud/guides/cloud-builds.md): Build and deploy agents on Pipecat Cloud without managing your own container registry, using the hosted build service. - [Pipecat Cloud Capacity Planning](https://docs.pipecat.ai/pipecat-cloud/guides/capacity-planning.md): Plan Pipecat Cloud capacity to minimize cold starts: auto-scaling behavior, the min-agents parameter, and sizing for call volume. #### Container Registries - [Pipecat Cloud Container Registries](https://docs.pipecat.ai/pipecat-cloud/guides/container-registries/overview.md): Deploy Pipecat Cloud agents from private container registries: supported registries, image pull credentials, and setup guides. - [Deploy from Docker Hub](https://docs.pipecat.ai/pipecat-cloud/guides/container-registries/docker-hub.md): Deploy Pipecat Cloud agents from Docker Hub: pushing your agent image, access tokens, and private repository configuration. - [Deploy from AWS ECR](https://docs.pipecat.ai/pipecat-cloud/guides/container-registries/aws-ecr.md): Deploy Pipecat Cloud agents from Amazon Elastic Container Registry: ECR authentication, image pull secrets, and configuration. - [Deploy from GCP Artifact Registry](https://docs.pipecat.ai/pipecat-cloud/guides/container-registries/gcp-artifact-registry.md): Deploy Pipecat Cloud agents from Google Cloud Artifact Registry: service account keys, image paths, and pull configuration. - [Pipecat Cloud CI with GitHub Actions](https://docs.pipecat.ai/pipecat-cloud/guides/ci-with-github-actions.md): Automate Pipecat Cloud builds and deploys from CI using the official pipecat-ai GitHub Action in your workflow. - [Pipecat Cloud Personal Access Tokens](https://docs.pipecat.ai/pipecat-cloud/guides/personal-access-tokens.md): Authenticate CI/CD pipelines and automation against Pipecat Cloud with non-interactive personal access tokens (PATs). - [Pipecat Cloud Daily WebRTC](https://docs.pipecat.ai/pipecat-cloud/guides/daily-webrtc.md): Use Daily as the WebRTC transport for Pipecat Cloud agents: room creation, tokens, and connecting clients to agent sessions. - [Krisp VIVA on Pipecat Cloud](https://docs.pipecat.ai/pipecat-cloud/guides/krisp-viva.md): Enable Krisp VIVA voice isolation on Pipecat Cloud to remove background noise and voices for more reliable agent conversations. #### Logging & Observability - [Datadog with Pipecat Cloud](https://docs.pipecat.ai/pipecat-cloud/guides/using-datadog.md): Ship Datadog logging, traces, and custom metrics from Pipecat Cloud agents for production monitoring and alerting. - [Pipecat Cloud Regions](https://docs.pipecat.ai/pipecat-cloud/guides/regions.md): Deploy Pipecat Cloud agents to geographic regions for lower latency and data residency: available regions and configuration. - [Pipecat Cloud Session API](https://docs.pipecat.ai/pipecat-cloud/guides/session-api.md): Send HTTP requests directly to running Pipecat Cloud agent sessions through the session API proxy for custom control. - [Smart Turn on Pipecat Cloud](https://docs.pipecat.ai/pipecat-cloud/guides/smart-turn.md): Use Pipecat's Smart Turn model on Pipecat Cloud for advanced conversational turn detection beyond VAD silence timeouts. #### WebSockets - [Pipecat Cloud Generic WebSocket](https://docs.pipecat.ai/pipecat-cloud/guides/generic-websocket.md): Connect any WebSocket client to Pipecat Cloud agents: telephony providers, server-to-server links, or custom integrations. - [Pipecat Cloud WebSocket Authentication](https://docs.pipecat.ai/pipecat-cloud/guides/websocket-authentication.md): Secure WebSocket connections to Pipecat Cloud agents with HMAC token authentication for telephony and custom clients. - [Pipecat Cloud WhatsApp Business Calling](https://docs.pipecat.ai/pipecat-cloud/guides/whatsapp.md): Receive WhatsApp voice calls in Pipecat Cloud agents with the WhatsApp Business Calling API: webhook setup and call flow. #### Telephony - [Pipecat Cloud Telephony](https://docs.pipecat.ai/pipecat-cloud/guides/telephony/overview.md): Connect telephony providers to Pipecat Cloud agents: Twilio, Telnyx, Plivo, Exotel WebSockets and Daily PSTN dial-in and dial-out. - [Twilio WebSockets on Pipecat Cloud](https://docs.pipecat.ai/pipecat-cloud/guides/telephony/twilio-websocket.md): Receive Twilio phone calls in Pipecat Cloud agents over Media Streams WebSockets: TwiML setup, serializers, and configuration. - [Telnyx WebSockets on Pipecat Cloud](https://docs.pipecat.ai/pipecat-cloud/guides/telephony/telnyx-websocket.md): Receive Telnyx phone calls in Pipecat Cloud agents over WebSocket media streaming: TeXML setup, serializers, and configuration. - [Plivo WebSockets on Pipecat Cloud](https://docs.pipecat.ai/pipecat-cloud/guides/telephony/plivo-websocket.md): Receive Plivo phone calls in Pipecat Cloud agents over Plivo's WebSocket transport: XML setup, serializers, and configuration. - [Exotel WebSockets on Pipecat Cloud](https://docs.pipecat.ai/pipecat-cloud/guides/telephony/exotel-websocket.md): Receive Exotel phone calls in Pipecat Cloud agents over Voicebot WebSocket streaming: applet setup, serializers, and config. - [Daily PSTN Dial-in on Pipecat Cloud](https://docs.pipecat.ai/pipecat-cloud/guides/telephony/daily-dial-in.md): Receive calls from PSTN numbers and SIP addresses in Pipecat Cloud agents using Daily's transport and dial-in configuration. - [Daily PSTN Dial-out on Pipecat Cloud](https://docs.pipecat.ai/pipecat-cloud/guides/telephony/daily-dial-out.md): Place outbound calls to PSTN numbers and SIP addresses from Pipecat Cloud agents using Daily's transport and dial-out config. ### Security & Compliance - [Pipecat Cloud Security Guide](https://docs.pipecat.ai/pipecat-cloud/security/security-and-compliance.md): Secure Pipecat Cloud agents: platform security model, compliance posture, and hardening recommendations for deployments. - [Pipecat Cloud HIPAA Compliance](https://docs.pipecat.ai/pipecat-cloud/security/hipaa.md): Run HIPAA-compliant voice agents on Pipecat Cloud: Business Associate Agreements, PHI handling, and compliance controls. ## API Reference ### Get Started - [API Reference](https://docs.pipecat.ai/api-reference/index.md): Reference documentation for every part of Pipecat: the Python server framework, client SDKs, Pipecat Flows, Pipecat Cloud, the CLI, etc. ### Pipecat Server - [Pipecat Server API Overview](https://docs.pipecat.ai/api-reference/server/introduction.md): API reference for the Pipecat Python framework: pipelines, services, transports, utilities, and the server runtime. #### Services - [Supported Services](https://docs.pipecat.ai/api-reference/server/services/supported-services.md): The full catalog of AI services integrated with Pipecat: STT, TTS, LLM, transports, and more, with setup requirements. - [Community Integrations](https://docs.pipecat.ai/api-reference/server/services/community-integrations.md): Community-maintained service integrations for Pipecat: how to find, install, and contribute them. ##### Transport - [Beyond Presence Transport](https://docs.pipecat.ai/api-reference/server/services/transport/beyond-presence.md): BeyTransport generates real-time Beyond Presence video avatars for Pipecat agents, built on Daily's WebRTC infrastructure. - [Daily WebRTC Transport](https://docs.pipecat.ai/api-reference/server/services/transport/daily.md): DailyTransport connects Pipecat pipelines to Daily's hosted WebRTC infrastructure for real-time audio and video sessions. - [FastAPI WebSocket Transport](https://docs.pipecat.ai/api-reference/server/services/transport/fastapi-websocket.md): FastAPIWebsocketTransport serves Pipecat sessions over WebSockets in FastAPI apps, pairing with telephony FrameSerializers. - [HeyGen Transport](https://docs.pipecat.ai/api-reference/server/services/transport/heygen.md): HeyGenTransport joins your Pipecat bot to the same virtual room as a HeyGen avatar for interactive avatar sessions. - [LemonSlice Transport](https://docs.pipecat.ai/api-reference/server/services/transport/lemonslice.md): LemonSliceTransport joins your Pipecat bot to a room with a LemonSlice avatar, layered on the Daily transport. - [LiveKit Transport](https://docs.pipecat.ai/api-reference/server/services/transport/livekit.md): LiveKitTransport provides real-time audio communication for Pipecat pipelines over LiveKit's WebRTC infrastructure. - [Small WebRTC Transport](https://docs.pipecat.ai/api-reference/server/services/transport/small-webrtc.md): SmallWebRTCTransport makes serverless peer-to-peer WebRTC connections between a client and your Pipecat bot, no provider needed. - [Tavus Transport](https://docs.pipecat.ai/api-reference/server/services/transport/tavus.md): TavusTransport joins your Pipecat bot to the same virtual room as a Tavus avatar for conversational video sessions. - [Vonage Video Connector Transport](https://docs.pipecat.ai/api-reference/server/services/transport/vonage.md): VonageVideoConnectorTransport provides real-time WebRTC audio and video for Pipecat using the Vonage Video API. - [WebSocket Transports](https://docs.pipecat.ai/api-reference/server/services/transport/websocket-server.md): WebsocketServerTransport and WebsocketClientTransport connect Pipecat to clients over plain WebSockets for real-time audio. - [WhatsApp Transport](https://docs.pipecat.ai/api-reference/server/services/transport/whatsapp.md): WhatsAppTransport connects Pipecat bots to WhatsApp voice calls via the WhatsApp Business API and SmallWebRTCTransport. - [TransportParams](https://docs.pipecat.ai/api-reference/server/services/transport/transport-params.md): TransportParams is the base configuration class shared by all Pipecat transports: audio, video, and VAD settings. ##### Serializers - [Frame Serializer Overview](https://docs.pipecat.ai/api-reference/server/services/serializers/introduction.md): How Pipecat FrameSerializers convert between internal frames and provider wire formats, and how to write a custom serializer. - [Acefone Frame Serializer](https://docs.pipecat.ai/api-reference/server/services/serializers/acefone.md): AcefoneFrameSerializer converts between Pipecat frames and Acefone's telephony WebSocket protocol for hosted voice agents. - [Asterisk Frame Serializer](https://docs.pipecat.ai/api-reference/server/services/serializers/asterisk.md): AsteriskFrameSerializer converts between Pipecat frames and an Asterisk server's media stream for self-hosted telephony. - [AwaazAI Frame Serializer](https://docs.pipecat.ai/api-reference/server/services/serializers/awaazai.md): AwaazAIFrameSerializer converts between Pipecat frames and Awaaz AI's telephony WebSocket streaming protocol. - [Bandwidth Frame Serializer](https://docs.pipecat.ai/api-reference/server/services/serializers/bandwidth.md): BandwidthFrameSerializer converts between Pipecat frames and Bandwidth Programmable Voice WebSocket media streams. - [Exotel Frame Serializer](https://docs.pipecat.ai/api-reference/server/services/serializers/exotel.md): ExotelFrameSerializer integrates Pipecat pipelines with Exotel's WebSocket media streaming protocol for phone calls. - [Genesys AudioHook Serializer](https://docs.pipecat.ai/api-reference/server/services/serializers/genesys.md): GenesysAudioHookSerializer connects Pipecat to Genesys Cloud Contact Center via the AudioHook WebSocket protocol. - [Plivo Frame Serializer](https://docs.pipecat.ai/api-reference/server/services/serializers/plivo.md): PlivoFrameSerializer integrates Pipecat with Plivo's Audio Streaming WebSocket protocol for telephony bots. - [Twilio Frame Serializer](https://docs.pipecat.ai/api-reference/server/services/serializers/twilio.md): TwilioFrameSerializer integrates Pipecat with Twilio's Media Streams WebSocket protocol for phone call audio. - [Telnyx Frame Serializer](https://docs.pipecat.ai/api-reference/server/services/serializers/telnyx.md): TelnyxFrameSerializer integrates Pipecat with Telnyx's WebSocket media streaming protocol for telephony bots. - [Vonage Frame Serializer](https://docs.pipecat.ai/api-reference/server/services/serializers/vonage.md): VonageFrameSerializer integrates Pipecat with the Vonage Video API Audio Connector WebSocket protocol. - [Wavix Frame Serializer](https://docs.pipecat.ai/api-reference/server/services/serializers/wavix.md): WavixFrameSerializer converts audio between Pipecat frames and the Wavix WebSocket media streaming protocol. ##### Speech-to-Text - [AssemblyAI Speech-to-Text](https://docs.pipecat.ai/api-reference/server/services/stt/assemblyai.md): Real-time STT with AssemblyAISTTService: WebSocket streaming, interim results, and end-of-turn detection for voice agents. - [AWS Transcribe Speech-to-Text](https://docs.pipecat.ai/api-reference/server/services/stt/aws.md): AWSTranscribeSTTService streams real-time STT over Amazon Transcribe's WebSocket API, with interim results and multi-language support. - [Azure Speech-to-Text](https://docs.pipecat.ai/api-reference/server/services/stt/azure.md): AzureSTTService runs continuous real-time STT on the Azure Cognitive Services Speech SDK, with custom endpoints and broad language support. - [Cartesia Speech-to-Text](https://docs.pipecat.ai/api-reference/server/services/stt/cartesia.md): Real-time STT with CartesiaSTTService and turn-aware CartesiaTurnsSTTService over Cartesia's streaming transcription API. - [Deepgram Speech-to-Text](https://docs.pipecat.ai/api-reference/server/services/stt/deepgram.md): Streaming STT with DeepgramSTTService and DeepgramFluxSTTService: Nova models, Flux turn detection, and SageMaker variants. - [ElevenLabs Speech-to-Text](https://docs.pipecat.ai/api-reference/server/services/stt/elevenlabs.md): Speech recognition with ElevenLabsRealtimeSTTService (live WebSocket streaming) and ElevenLabsSTTService for file-based STT. - [Fal Wizper Speech-to-Text](https://docs.pipecat.ai/api-reference/server/services/stt/fal.md): FalSTTService transcribes speech segments with Fal's Wizper API, using VAD-segmented audio to cut STT cost and latency. - [Floe Speech-to-Text](https://docs.pipecat.ai/api-reference/server/services/stt/floe.md): Streaming speech-to-text service over Floe's transcription WebSocket with per-call spend caps - [FunASR Speech-to-Text](https://docs.pipecat.ai/api-reference/server/services/stt/funasr.md): Offline STT with FunASRSTTService running SenseVoice models locally on CPU or GPU: multilingual, with strong Chinese support. - [Gladia Speech-to-Text](https://docs.pipecat.ai/api-reference/server/services/stt/gladia.md): GladiaSTTService streams real-time STT over Gladia's WebSocket API: 99+ languages, custom vocabulary, translation, and sentiment. - [Gnani Speech-to-Text](https://docs.pipecat.ai/api-reference/server/services/stt/gnani.md): STT for 10+ Indian languages with GnaniSTTService (WebSocket streaming) and GnaniHttpSTTService (REST) on the Gnani platform. - [Google Speech-to-Text](https://docs.pipecat.ai/api-reference/server/services/stt/google.md): GoogleSTTService provides real-time STT on Google Cloud Speech-to-Text V2, with 125+ languages, model selection, and VAD. - [Gradium Speech-to-Text](https://docs.pipecat.ai/api-reference/server/services/stt/gradium.md): GradiumSTTService streams real-time STT over Gradium's WebSocket API with multilingual transcription and semantic VAD. - [Groq Whisper Speech-to-Text](https://docs.pipecat.ai/api-reference/server/services/stt/groq.md): GroqSTTService runs VAD-segmented STT on Groq's hosted Whisper API, pairing Whisper accuracy with ultra-fast inference. - [Hakim Speech-to-Text](https://docs.pipecat.ai/api-reference/server/services/stt/hakim.md): HakimSTTService provides Arabic-first realtime STT over a persistent WebSocket connection to Hakim's transcription API. - [Mistral Speech-to-Text](https://docs.pipecat.ai/api-reference/server/services/stt/mistral.md): MistralSTTService streams real-time STT through Mistral's Voxtral Realtime API via the Mistral SDK's RealtimeConnection. - [Moonshine Speech-to-Text](https://docs.pipecat.ai/api-reference/server/services/stt/moonshine.md): Offline STT with MoonshineSTTService: small, fast Moonshine ASR models running locally on CPU via ONNX Runtime, no GPU or API key. - [NVIDIA Nemotron Speech-to-Text](https://docs.pipecat.ai/api-reference/server/services/stt/nvidia.md): Real-time STT with NvidiaSTTService, NvidiaSegmentedSTTService, and a SageMaker variant, running NVIDIA Nemotron Speech models. - [OpenAI Speech-to-Text](https://docs.pipecat.ai/api-reference/server/services/stt/openai.md): STT with OpenAIRealtimeSTTService for live streaming and OpenAISTTService for segmented transcription of OpenAI speech models. - [Quickdial Speech-to-Text](https://docs.pipecat.ai/api-reference/server/services/stt/quickdial.md): QuickdialSTTService extends SegmentedSTTService to transcribe speech with Quickdial's whisper.cpp-powered STT API. - [Ringg AI Speech-to-Text](https://docs.pipecat.ai/api-reference/server/services/stt/ringg.md): RinggSTTService streams STT through the official ringglabs Python SDK, which manages the WebSocket connection and event parsing. - [Sarvam AI Speech-to-Text](https://docs.pipecat.ai/api-reference/server/services/stt/sarvam.md): SarvamSTTService streams real-time STT for Indian languages over Sarvam AI's WebSocket API, with VAD-based segmentation. - [Simplismart Speech-to-Text](https://docs.pipecat.ai/api-reference/server/services/stt/simplismart.md): SimplismartSTTService POSTs WAV audio segments to Simplismart's HTTP /predict endpoint for segmented STT transcription. - [SLNG Speech-to-Text](https://docs.pipecat.ai/api-reference/server/services/stt/slng.md): SlngSTTService streams real-time STT over a persistent WebSocket to SLNG, a unified voice AI gateway routing multiple providers. - [Smallest AI Speech-to-Text](https://docs.pipecat.ai/api-reference/server/services/stt/smallest.md): SmallestSTTService provides real-time streaming STT through Smallest AI's Pulse model over the Waves WebSocket API. - [Soniox Speech-to-Text](https://docs.pipecat.ai/api-reference/server/services/stt/soniox.md): SonioxSTTService streams real-time STT over Soniox's WebSocket API: 60+ languages, custom context, and mixed-language audio. - [Speechmatics Speech-to-Text](https://docs.pipecat.ai/api-reference/server/services/stt/speechmatics.md): SpeechmaticsSTTService delivers real-time STT with partial and final results, speaker diarization, and end-of-utterance detection. - [Together AI Speech-to-Text](https://docs.pipecat.ai/api-reference/server/services/stt/together.md): TogetherSTTService streams real-time STT via Together AI's OpenAI-compatible WebSocket transcription endpoints. - [Uplift AI Speech-to-Text](https://docs.pipecat.ai/api-reference/server/services/stt/upliftai.md): UpliftAISTTService transcribes Urdu-optimized speech using Uplift AI's Scribe STT API with VAD-segmented audio. - [Whisper Speech-to-Text](https://docs.pipecat.ai/api-reference/server/services/stt/whisper.md): Offline STT with WhisperSTTService: OpenAI Whisper models running locally, with multiple sizes and hardware acceleration. - [xAI Speech-to-Text](https://docs.pipecat.ai/api-reference/server/services/stt/xai.md): XAISTTService streams real-time STT over xAI's WebSocket API with interim results, configurable endpointing, and multichannel audio. ##### LLM - [Anannas AI LLM](https://docs.pipecat.ai/api-reference/server/services/llm/anannas.md): AnannasLLMService routes chat completions through Anannas AI's unified gateway: 500+ models behind one OpenAI-compatible API. - [Anthropic LLM](https://docs.pipecat.ai/api-reference/server/services/llm/anthropic.md): AnthropicLLMService integrates Claude models with streaming responses, function calling, and prompt caching support. - [AWS Bedrock LLM](https://docs.pipecat.ai/api-reference/server/services/llm/aws.md): AWSBedrockLLMService runs Amazon Bedrock foundation models, including Claude and Amazon Nova, with streaming and function calling. - [Azure OpenAI LLM](https://docs.pipecat.ai/api-reference/server/services/llm/azure.md): AzureLLMService reaches Azure OpenAI language models through the OpenAI-compatible interface it inherits from OpenAILLMService. - [Baseten LLM](https://docs.pipecat.ai/api-reference/server/services/llm/baseten.md): BasetenLLMService connects to Baseten Model APIs and dedicated deployments, inheriting OpenAILLMService's streaming and tools. - [Cerebras LLM](https://docs.pipecat.ai/api-reference/server/services/llm/cerebras.md): CerebrasLLMService accesses Cerebras language models for chat completion via an OpenAI-compatible interface with streaming. - [Crusoe LLM](https://docs.pipecat.ai/api-reference/server/services/llm/crusoe.md): CrusoeLLMService generates chat completions on Crusoe Cloud's Managed Inference API, OpenAI-compatible with streaming support. - [DeepSeek LLM](https://docs.pipecat.ai/api-reference/server/services/llm/deepseek.md): DeepSeekLLMService accesses DeepSeek language models through an OpenAI-compatible interface, inheriting from OpenAILLMService. - [Fireworks AI LLM](https://docs.pipecat.ai/api-reference/server/services/llm/fireworks.md): FireworksLLMService accesses Fireworks AI language models with streaming and function calling via an OpenAI-compatible API. - [Floe LLM](https://docs.pipecat.ai/api-reference/server/services/llm/floe.md): LLM service that routes OpenAI-compatible chat completions through Floe with per-call spend caps - [Google Gemini LLM](https://docs.pipecat.ai/api-reference/server/services/llm/google.md): GoogleLLMService integrates Google's Gemini models with streaming responses, function calling, and multimodal input support. - [Google Vertex AI LLM](https://docs.pipecat.ai/api-reference/server/services/llm/google-vertex.md): GoogleVertexLLMService extends GoogleLLMService with Vertex AI authentication for running Gemini models on Google Cloud. - [Grok LLM](https://docs.pipecat.ai/api-reference/server/services/llm/grok.md): GrokLLMService accesses xAI's Grok language models through an OpenAI-compatible interface with streaming and function calling. - [Groq LLM](https://docs.pipecat.ai/api-reference/server/services/llm/groq.md): GroqLLMService runs chat completions on Groq's high-speed inference API through an OpenAI-compatible interface. - [Inception LLM](https://docs.pipecat.ai/api-reference/server/services/llm/inception.md): InceptionLLMService serves Inception's Mercury-2 diffusion-based reasoning model via an OpenAI-compatible interface. - [Mistral LLM](https://docs.pipecat.ai/api-reference/server/services/llm/mistral.md): MistralLLMService accesses Mistral language models through an OpenAI-compatible interface, inheriting OpenAILLMService streaming. - [Nebius LLM](https://docs.pipecat.ai/api-reference/server/services/llm/nebius.md): NebiusLLMService generates chat completions on Nebius Token Factory's OpenAI-compatible API with streaming responses. - [Novita AI LLM](https://docs.pipecat.ai/api-reference/server/services/llm/novita.md): NovitaLLMService connects Pipecat to Novita AI's language model catalog through an OpenAI-compatible streaming interface. - [NVIDIA NIM LLM](https://docs.pipecat.ai/api-reference/server/services/llm/nvidia.md): NvidiaLLMService accesses NVIDIA NIM (NVIDIA Inference Microservice) models via an OpenAI-compatible interface. - [Ollama LLM](https://docs.pipecat.ai/api-reference/server/services/llm/ollama.md): OLLamaLLMService runs locally hosted Ollama models with no cloud API, via BaseOpenAILLMService's OpenAI-compatible interface. - [OpenAI LLM](https://docs.pipecat.ai/api-reference/server/services/llm/openai.md): OpenAILLMService for GPT chat completions: streaming responses, function calling, vision input, and OpenAI-compatible base URLs. - [OpenAI Responses LLM](https://docs.pipecat.ai/api-reference/server/services/llm/openai-responses.md): OpenAIResponsesLLMService and OpenAIResponsesHttpLLMService bring OpenAI's Responses API to Pipecat voice pipelines. - [OpenRouter LLM](https://docs.pipecat.ai/api-reference/server/services/llm/openrouter.md): OpenRouterLLMService reaches OpenRouter's multi-provider model catalog through an OpenAI-compatible streaming interface. - [Perplexity LLM](https://docs.pipecat.ai/api-reference/server/services/llm/perplexity.md): PerplexityLLMService accesses Perplexity language models through an OpenAI-compatible interface, inheriting OpenAILLMService. - [Qwen LLM](https://docs.pipecat.ai/api-reference/server/services/llm/qwen.md): QwenLLMService runs Alibaba Cloud's Qwen models through an OpenAI-compatible interface with streaming chat completions. - [SambaNova LLM](https://docs.pipecat.ai/api-reference/server/services/llm/sambanova.md): SambaNovaLLMService connects to SambaNova language models through an OpenAI-compatible interface with streaming support. - [Sarvam AI LLM](https://docs.pipecat.ai/api-reference/server/services/llm/sarvam.md): SarvamLLMService generates chat completions on Sarvam AI's OpenAI-compatible API with streaming and function calling. - [Simplismart LLM](https://docs.pipecat.ai/api-reference/server/services/llm/simplismart.md): SimplismartLLMService generates chat completions with Simplismart's OpenAI-compatible Chat Completions API, extending BaseOpenAILLMService. - [Together AI LLM](https://docs.pipecat.ai/api-reference/server/services/llm/together.md): TogetherLLMService accesses Together AI's model catalog, including Meta Llama 3.1 and 3.2, via an OpenAI-compatible interface. ##### Text-to-Speech - [AsyncAI Text-to-Speech](https://docs.pipecat.ai/api-reference/server/services/tts/asyncai.md): TTS with AsyncAITTSService for real-time WebSocket streaming with interruption support, and AsyncAIHttpTTSService for HTTP synthesis. - [AWS Polly Text-to-Speech](https://docs.pipecat.ai/api-reference/server/services/tts/aws.md): AWSPollyTTSService synthesizes TTS through Amazon Polly's standard, neural, and generative engines with configurable voice options. - [Azure Text-to-Speech](https://docs.pipecat.ai/api-reference/server/services/tts/azure.md): TTS on Azure Cognitive Services: AzureTTSService for real-time WebSocket streaming and AzureHttpTTSService for simpler HTTP synthesis. - [Camb AI Text-to-Speech](https://docs.pipecat.ai/api-reference/server/services/tts/camb.md): CambTTSService streams high-quality TTS using Camb AI's MARS model family, with multiple model and voice options. - [Cartesia Text-to-Speech](https://docs.pipecat.ai/api-reference/server/services/tts/cartesia.md): Low-latency TTS with CartesiaTTSService (WebSocket streaming, word timestamps) and CartesiaHttpTTSService for batch synthesis. - [Deepdub Text-to-Speech](https://docs.pipecat.ai/api-reference/server/services/tts/deepdub.md): DeepdubTTSService streams real-time TTS over Deepdub AI's WebSocket API, extending InterruptibleTTSService for interruption handling. - [Deepgram Text-to-Speech](https://docs.pipecat.ai/api-reference/server/services/tts/deepgram.md): Aura TTS with DeepgramTTSService, DeepgramHttpTTSService, DeepgramFluxTTSService, and a SageMaker variant for hosted inference. - [ElevenLabs Text-to-Speech](https://docs.pipecat.ai/api-reference/server/services/tts/elevenlabs.md): TTS with ElevenLabsTTSService (WebSocket streaming, word-level timestamps) and ElevenLabsHttpTTSService for HTTP synthesis. - [Fish Audio Text-to-Speech](https://docs.pipecat.ai/api-reference/server/services/tts/fish.md): FishAudioTTSService streams real-time TTS through Fish Audio's WebSocket API with custom voice models and prosody control. - [Floe Text-to-Speech](https://docs.pipecat.ai/api-reference/server/services/tts/floe.md): Text-to-speech service that routes OpenAI-compatible speech synthesis through Floe with per-call spend caps - [Gnani Text-to-Speech](https://docs.pipecat.ai/api-reference/server/services/tts/gnani.md): TTS for 10+ Indian languages with GnaniTTSService (WebSocket), GnaniSSETTSService (SSE), and GnaniHttpTTSService (REST). - [Google Text-to-Speech](https://docs.pipecat.ai/api-reference/server/services/tts/google.md): Google Cloud TTS with GoogleTTSService (low-latency streaming), GoogleHttpTTSService, and GeminiTTSService for Gemini TTS models. - [Gradium Text-to-Speech](https://docs.pipecat.ai/api-reference/server/services/tts/gradium.md): GradiumTTSService streams expressive TTS over Gradium's WebSocket API with instant voice cloning and low-latency inference. - [Groq Text-to-Speech](https://docs.pipecat.ai/api-reference/server/services/tts/groq.md): GroqTTSService delivers fast TTS from Groq's API with multiple voice options at a fixed 48kHz sample rate for voice agents. - [Hakim Text-to-Speech](https://docs.pipecat.ai/api-reference/server/services/tts/hakim.md): HakimTTSService synthesizes Arabic-first realtime TTS, subclassing InterruptibleTTSService for interruption handling. - [Hume Text-to-Speech](https://docs.pipecat.ai/api-reference/server/services/tts/hume.md): HumeTTSService generates expressive TTS with Hume AI's Octave models: context-aware pitch, speed, emotion, and word timestamps. - [Inworld Text-to-Speech](https://docs.pipecat.ai/api-reference/server/services/tts/inworld.md): TTS with InworldTTSService (real-time WebSocket) and InworldHttpTTSService, running Inworld AI's TTS-2 and TTS-1.5 models. - [Kokoro Text-to-Speech](https://docs.pipecat.ai/api-reference/server/services/tts/kokoro.md): Local offline TTS with KokoroTTSService and the kokoro-onnx engine: on-device synthesis with no external API calls. - [LMNT Text-to-Speech](https://docs.pipecat.ai/api-reference/server/services/tts/lmnt.md): LMNTTTSService streams TTS through LMNT's WebSocket API, optimized for ultra-low-latency conversational voice agents. - [Lokutor Text-to-Speech](https://docs.pipecat.ai/api-reference/server/services/tts/lokutor.md): LokutorTTSService generates TTS over a persistent WebSocket connection to Lokutor's CPU-only speech synthesis API. - [MiniMax Text-to-Speech](https://docs.pipecat.ai/api-reference/server/services/tts/minimax.md): TTS with MiniMaxTTSService and MiniMaxHttpTTSService on MiniMax's T2A API: streaming, emotional voice control, and voice options. - [Mistral Text-to-Speech](https://docs.pipecat.ai/api-reference/server/services/tts/mistral.md): MistralTTSService synthesizes speech with Mistral's Voxtral TTS API, streaming PCM audio over HTTP Server-Sent Events. - [Murf AI Text-to-Speech](https://docs.pipecat.ai/api-reference/server/services/tts/murf.md): MurfTTSService streams real-time TTS from Murf AI's WebSocket API with voice customization and styling options. - [Neuphonic Text-to-Speech](https://docs.pipecat.ai/api-reference/server/services/tts/neuphonic.md): TTS with NeuphonicTTSService (real-time WebSocket streaming with interruptions) and NeuphonicHttpTTSService for HTTP synthesis. - [NVIDIA Nemotron Text-to-Speech](https://docs.pipecat.ai/api-reference/server/services/tts/nvidia.md): TTS with NvidiaTTSService plus SageMaker variants, running NVIDIA Nemotron Speech models for real-time synthesis. - [OpenAI Text-to-Speech](https://docs.pipecat.ai/api-reference/server/services/tts/openai.md): OpenAITTSService provides TTS through OpenAI's speech API, supporting both traditional and GPT-4o family voice models. - [Piper Text-to-Speech](https://docs.pipecat.ai/api-reference/server/services/tts/piper.md): PiperTTSService and PiperHttpTTSService synthesize neural TTS from a self-hosted Piper server: private, no external API. - [Pocket TTS](https://docs.pipecat.ai/api-reference/server/services/tts/pocket-tts.md): PocketTTSService runs kyutai-labs' pocket-tts streaming model for local, CPU-only TTS with no external API or GPU. - [Resemble AI Text-to-Speech](https://docs.pipecat.ai/api-reference/server/services/tts/resembleai.md): ResembleAITTSService streams TTS over Resemble AI's WebSocket API with word-level timestamps and audio context management. - [Respeecher Text-to-Speech](https://docs.pipecat.ai/api-reference/server/services/tts/respeecher.md): RespeecherTTSService streams real-time TTS audio over a WebSocket connection to Respeecher's voice synthesis API. - [Quickdial Text-to-Speech](https://docs.pipecat.ai/api-reference/server/services/tts/quickdial.md): QuickdialTTSService generates speech from Quickdial's real-time TTS API, extending Pipecat's base TTSService. - [Rime Text-to-Speech](https://docs.pipecat.ai/api-reference/server/services/tts/rime.md): TTS with RimeTTSService (WebSocket streaming with word timing and interruptions) and RimeHttpTTSService for batch synthesis. - [Rumik AI Text-to-Speech](https://docs.pipecat.ai/api-reference/server/services/tts/rumik.md): pipecat-rumik provides RumikTTSService (WebSocket streaming) and RumikHttpTTSService for TTS with Rumik AI's APIs. - [Sarvam AI Text-to-Speech](https://docs.pipecat.ai/api-reference/server/services/tts/sarvam.md): TTS for Indian languages with SarvamTTSService and SarvamHttpTTSService, with extensive voice customization on Sarvam AI's API. - [Simplismart Text-to-Speech](https://docs.pipecat.ai/api-reference/server/services/tts/simplismart.md): SimplismartHttpTTSService streams raw PCM TTS audio from Simplismart's /tts API endpoint for low-latency synthesis. - [SLNG Text-to-Speech](https://docs.pipecat.ai/api-reference/server/services/tts/slng.md): SlngTTSService and SlngHttpTTSService synthesize TTS through SLNG, a unified voice AI gateway routing multiple providers. - [Smallest AI Text-to-Speech](https://docs.pipecat.ai/api-reference/server/services/tts/smallest.md): SmallestTTSService streams real-time TTS through Smallest AI's Waves WebSocket API with configurable voice parameters. - [Soniox Text-to-Speech](https://docs.pipecat.ai/api-reference/server/services/tts/soniox.md): SonioxTTSService streams text incrementally to Soniox's WebSocket TTS endpoint for real-time speech synthesis. - [Speechmatics Text-to-Speech](https://docs.pipecat.ai/api-reference/server/services/tts/speechmatics.md): SpeechmaticsTTSService provides production-grade low-latency TTS, streaming 16kHz mono audio tuned for telephony voice agents. - [Supertonic Text-to-Speech](https://docs.pipecat.ai/api-reference/server/services/tts/supertonic.md): SupertonicTTSService wraps the official Supertonic Python SDK as a Pipecat-compatible TTSService for speech synthesis. - [Together AI Text-to-Speech](https://docs.pipecat.ai/api-reference/server/services/tts/together.md): TogetherTTSService streams real-time TTS over Together AI's WebSocket API with configurable voice and model options. - [Pipecat TTS Cache](https://docs.pipecat.ai/api-reference/server/services/tts/tts-cache.md): TTSCacheMixin transparently wraps any Pipecat TTS service, caching repeated phrases to cut API costs and response latency. - [Typecast Text-to-Speech](https://docs.pipecat.ai/api-reference/server/services/tts/typecast.md): TypecastTTSService synthesizes expressive TTS with Typecast's neural voices, with emotion control and audio customization. - [Uplift AI Text-to-Speech](https://docs.pipecat.ai/api-reference/server/services/tts/upliftai.md): UpliftHttpTTSService converts text to speech with Uplift AI's HTTP TTS API for high-quality, low-latency synthesis. - [Voice.ai Text-to-Speech](https://docs.pipecat.ai/api-reference/server/services/tts/voiceai.md): VoiceAiTTSService streams TTS through Voice.ai's Multi-Context WebSocket API over a single persistent connection. - [xAI Text-to-Speech](https://docs.pipecat.ai/api-reference/server/services/tts/xai.md): TTS in 20 languages with XAITTSService (WebSocket streaming) and XAIHttpTTSService on xAI's speech synthesis APIs. - [Coqui XTTS Text-to-Speech](https://docs.pipecat.ai/api-reference/server/services/tts/xtts.md): XTTSTTSService provides multilingual TTS with voice cloning through a locally hosted Coqui XTTS streaming server. - [XTTS-vLLM Text-to-Speech](https://docs.pipecat.ai/api-reference/server/services/tts/xtts-vllm.md): XTTSVLLMTTSService streams TTS from a self-hosted XTTSv2 + vLLM streaming server for low-latency open-model synthesis. ##### Realtime LLM - [AWS Nova Sonic](https://docs.pipecat.ai/api-reference/server/services/s2s/aws.md): AWSNovaSonicLLMService provides real-time speech-to-speech (S2S) conversation with AWS's Nova Sonic model. - [Gemini Live](https://docs.pipecat.ai/api-reference/server/services/s2s/gemini-live.md): GeminiLiveLLMService connects Pipecat to Google's Gemini Live API for real-time speech-to-speech multimodal conversation. - [Gemini Live Vertex AI](https://docs.pipecat.ai/api-reference/server/services/s2s/gemini-live-vertex.md): GeminiLiveVertexLLMService runs Gemini Live speech-to-speech conversation through Vertex AI with Google Cloud authentication. - [Grok Realtime](https://docs.pipecat.ai/api-reference/server/services/s2s/grok.md): GrokRealtimeLLMService provides real-time multimodal speech-to-speech conversation using xAI's Grok Realtime API. - [Inworld Realtime](https://docs.pipecat.ai/api-reference/server/services/s2s/inworld.md): InworldRealtimeLLMService provides real-time multimodal conversation using Inworld's Realtime API for voice agents. - [OpenAI Realtime](https://docs.pipecat.ai/api-reference/server/services/s2s/openai.md): OpenAIRealtimeLLMService provides speech-to-speech conversation over the OpenAI Realtime API, replacing separate STT, LLM, and TTS. - [Ultravox Realtime](https://docs.pipecat.ai/api-reference/server/services/s2s/ultravox.md): UltravoxRealtimeLLMService provides real-time conversational AI with Ultravox Realtime, with agent and one-shot input modes. ##### Image Generation - [Azure OpenAI Image Generation](https://docs.pipecat.ai/api-reference/server/services/image-generation/azure.md): AzureImageGenServiceREST generates images with Azure OpenAI's image API from text prompts in Pipecat pipelines. - [fal Image Generation](https://docs.pipecat.ai/api-reference/server/services/image-generation/fal.md): FalImageGenService generates images from text prompts using fal's fast Stable Diffusion models in Pipecat pipelines. - [Google Image Generation](https://docs.pipecat.ai/api-reference/server/services/image-generation/google.md): GoogleImageGenService generates images with Google's Imagen models from text prompts in Pipecat pipelines. - [OpenAI Image Generation](https://docs.pipecat.ai/api-reference/server/services/image-generation/openai.md): OpenAIImageGenService generates images with OpenAI's DALL-E models from text prompts in Pipecat pipelines. - [Replicate Image Generation](https://docs.pipecat.ai/api-reference/server/services/image-generation/replicate.md): ReplicateImageGenService generates images from text prompts using models hosted on Replicate, within Pipecat pipelines. ##### Video - [Anam Video Avatar](https://docs.pipecat.ai/api-reference/server/services/video/anam.md): AnamVideoService generates real-time interactive Anam video avatars synchronized with your Pipecat agent's audio. - [HeyGen Video Avatar](https://docs.pipecat.ai/api-reference/server/services/video/heygen.md): HeyGenVideoService drives HeyGen LiveAvatar interactive video avatars from your Pipecat agent's speech output. - [Protoface Video Avatar](https://docs.pipecat.ai/api-reference/server/services/video/protoface.md): ProtofaceVideoService renders Pipecat TTS output through a real-time Protoface video avatar for conversational AI. - [Simli Video Avatar](https://docs.pipecat.ai/api-reference/server/services/video/simli.md): SimliVideoService creates real-time Simli AI avatar video over WebRTC, animated by your Pipecat agent's audio. - [Tavus Video Avatar](https://docs.pipecat.ai/api-reference/server/services/video/tavus.md): TavusVideoService generates Tavus AI avatar video that speaks your Pipecat agent's TTS output in real time. ##### Memory - [Mem0 Long-Term Memory](https://docs.pipecat.ai/api-reference/server/services/memory/mem0.md): Mem0MemoryService adds long-term conversation memory to Pipecat agents, recalling context across sessions with Mem0. - [Synap Memory](https://docs.pipecat.ai/api-reference/server/services/memory/synap.md): The synap-pipecat integration gives Pipecat voice agents persistent, cross-session memory backed by Synap. ##### Vision - [Moondream Vision](https://docs.pipecat.ai/api-reference/server/services/vision/moondream.md): MoondreamService runs local image analysis and visual question answering with the Moondream vision model. - [SmolVLM Vision](https://docs.pipecat.ai/api-reference/server/services/vision/smolvlm.md): SmolVlmService runs HuggingFace SmolVLM vision-language models locally for image analysis in Pipecat pipelines. ##### Analytics & Monitoring - [Finchvox Session Replay](https://docs.pipecat.ai/api-reference/server/services/analytics/finchvox.md): Finchvox is local session replay for voice AI: unified conversation timelines, audio, and traces for Pipecat debugging. - [Future AGI Observability](https://docs.pipecat.ai/api-reference/server/services/analytics/future-agi.md): traceAI-pipecat adds OpenTelemetry tracing and evaluation instrumentation from Future AGI to Pipecat applications. - [Noveum Trace](https://docs.pipecat.ai/api-reference/server/services/analytics/noveum-trace.md): Community-maintained observability integration that maps Pipecat pipeline events into Noveum traces and spans - [OpenInference Tracing](https://docs.pipecat.ai/api-reference/server/services/analytics/openinference.md): OpenTelemetry-based auto-instrumentation that converts Pipecat traces to OpenInference for viewing in Arize Phoenix or Arize AX - [Roark Call Analytics](https://docs.pipecat.ai/api-reference/server/services/analytics/roark.md): RoarkObserver sends Pipecat call analytics to Roark: call lifecycle, transcripts, tool calls, and recordings from one observer. - [Sentry Metrics](https://docs.pipecat.ai/api-reference/server/services/analytics/sentry.md): SentryMetrics extends FrameProcessorMetrics for performance monitoring of Pipecat pipelines with Sentry. ##### Knowledge Retrieval - [Moss Knowledge Retrieval](https://docs.pipecat.ai/api-reference/server/services/knowledge-retrieval/moss.md): MossRetrievalService injects low-latency semantic search results from Moss into your Pipecat agent's context. ##### Translation - [Pinch Translation](https://docs.pipecat.ai/api-reference/server/services/translation/pinch.md): PinchTranslatorService provides real-time speech-to-speech translation in Pipecat pipelines, powered by Pinch. ##### Audio Filters - [AIC Audio Filter](https://docs.pipecat.ai/api-reference/server/services/audio-filters/aic-filter.md): AICFilter enhances user speech with the ai-coustics SDK, reducing background noise in real-time Pipecat audio. - [Arctan Audio Filter](https://docs.pipecat.ai/api-reference/server/services/audio-filters/arctan.md): ArctanAudioFilter provides real-time speech enhancement and noise cancellation with Arctan Eigen models. - [Koala Audio Filter](https://docs.pipecat.ai/api-reference/server/services/audio-filters/koala-filter.md): KoalaFilter reduces background noise in real-time audio using Picovoice Koala noise suppression in Pipecat pipelines. - [Krisp VIVA Audio Filter](https://docs.pipecat.ai/api-reference/server/services/audio-filters/krisp-viva-filter.md): KrispVivaFilter isolates the user's voice in real-time audio using the Krisp VIVA model in Pipecat pipelines. - [RNNoise Audio Filter](https://docs.pipecat.ai/api-reference/server/services/audio-filters/rnnoise-filter.md): RNNoiseFilter suppresses background noise in real-time audio with the RNNoise recurrent neural network. ##### VAD - [AIC Quail VAD](https://docs.pipecat.ai/api-reference/server/services/vad/aic-quail-vad-analyzer.md): AICQuailVADAnalyzer is a standalone voice activity detection (VAD) analyzer powered by ai-coustics Quail. - [FireRed VAD](https://docs.pipecat.ai/api-reference/server/services/vad/fire-vad.md): FireVadAnalyzer provides streaming voice activity detection (VAD) for Pipecat using the FireRedVAD model. - [Krisp VIVA VAD](https://docs.pipecat.ai/api-reference/server/services/vad/krisp-viva-vad-analyzer.md): KrispVivaVadAnalyzer detects voice activity (VAD) using the Krisp VIVA SDK for accurate speech boundaries. - [Silero VAD](https://docs.pipecat.ai/api-reference/server/services/vad/silero-vad-analyzer.md): SileroVADAnalyzer detects voice activity with the Silero VAD ONNX model for speech detection in Pipecat pipelines. - [TEN VAD](https://docs.pipecat.ai/api-reference/server/services/vad/ten-vad.md): TenVadAnalyzer implements Pipecat's VADAnalyzer interface with the TEN VAD backend for voice activity detection. #### Extensions - [IVRNavigator](https://docs.pipecat.ai/api-reference/server/extensions/ivr.md): AI-powered Interactive Voice Response system navigation with automatic classification and goal-oriented decision making - [Pipecat Backchannel](https://docs.pipecat.ai/api-reference/server/extensions/pipecat-backchannel.md): Backchannel implementation that plays short listening sounds while the user is still speaking - [VoicemailDetector](https://docs.pipecat.ai/api-reference/server/extensions/voicemail.md): Technical implementation details and architecture for voicemail detection #### Utilities ##### Advanced Frame Processors - [LLMTextProcessor](https://docs.pipecat.ai/api-reference/server/utilities/frame/llm-text-processor.md): A processor for aggregating LLMTextFrames into logical units before passing them to downstream services - [Producer & Consumer Processors](https://docs.pipecat.ai/api-reference/server/utilities/frame/producer-consumer.md): Route frames between different parts of a pipeline, allowing selective frame sharing across parallel branches or within complex pipelines ##### Audio Processing - [AudioBufferProcessor](https://docs.pipecat.ai/api-reference/server/utilities/audio/audio-buffer-processor.md): Process and buffer audio frames from conversations with flexible event handling - [SoundfileMixer](https://docs.pipecat.ai/api-reference/server/utilities/audio/soundfile-mixer.md): SoundfileMixer combines real-time Pipecat audio with sound files for background audio and effects. ##### Context - [LLMContextSummarizer](https://docs.pipecat.ai/api-reference/server/utilities/context-summarization.md): Reference for LLMContextSummarizer, LLMAutoContextSummarizationConfig, LLMContextSummaryConfig, and SummaryAppliedEvent. ##### Frame Filters - [FrameFilter](https://docs.pipecat.ai/api-reference/server/utilities/filters/frame-filter.md): FrameFilter selectively passes only specified frame types through a Pipecat pipeline, dropping the rest. - [FunctionFilter](https://docs.pipecat.ai/api-reference/server/utilities/filters/function-filter.md): FunctionFilter filters frames in a Pipecat pipeline using a custom filter function for flexible frame control. - [IdentityFilter](https://docs.pipecat.ai/api-reference/server/utilities/filters/identify-filter.md): IdentityFilter passes all frames through unchanged, a no-op Pipecat processor useful for testing pipelines. - [NullFilter](https://docs.pipecat.ai/api-reference/server/utilities/filters/null-filter.md): NullFilter blocks every frame except system frames, useful for muting a Pipecat pipeline branch. - [WakeCheckFilter](https://docs.pipecat.ai/api-reference/server/utilities/filters/wake-check-filter.md): Processor that passes frames only after detecting wake phrases in transcriptions - [WakeNotifierFilter](https://docs.pipecat.ai/api-reference/server/utilities/filters/wake-notifier-filter.md): Processor that triggers a notifier when specified frame types pass a custom filter ##### Metrics and Telemetry - [OpenTelemetry Tracing](https://docs.pipecat.ai/api-reference/server/utilities/opentelemetry.md): Monitor and analyze your Pipecat conversational pipelines using OpenTelemetry ##### MCP - [MCPClient](https://docs.pipecat.ai/api-reference/server/utilities/mcp/mcp.md): MCPClient connects Pipecat bots to MCP (Model Context Protocol) servers and registers their tools with your LLM. ##### Observers - [Observer Pattern](https://docs.pipecat.ai/api-reference/server/utilities/observers/observer-pattern.md): The Pipecat observer pattern: watch frames flow through a pipeline without modifying them, and write your own observers. - [Debug Log Observer](https://docs.pipecat.ai/api-reference/server/utilities/observers/debug-observer.md): DebugLogObserver logs every frame flowing through a Pipecat pipeline with configurable filtering by frame type. - [LLM Log Observer](https://docs.pipecat.ai/api-reference/server/utilities/observers/llm-observer.md): LLMLogObserver logs LLM activity in a Pipecat pipeline: context frames, responses, and function calls. - [Transcription Log Observer](https://docs.pipecat.ai/api-reference/server/utilities/observers/transcription-observer.md): TranscriptionLogObserver logs speech-to-text transcription activity in a Pipecat pipeline for STT debugging. - [Startup Timing Observer](https://docs.pipecat.ai/api-reference/server/utilities/observers/startup-timing-observer.md): Measure processor startup times and transport readiness during pipeline initialization - [Turn Tracking Observer](https://docs.pipecat.ai/api-reference/server/utilities/observers/turn-tracking-observer.md): TurnTrackingObserver tracks conversation turns in a Pipecat pipeline and emits turn started and ended events. - [User-Bot Latency Observer](https://docs.pipecat.ai/api-reference/server/utilities/observers/user-bot-latency-observer.md): Measure response time between user speech and bot responses in Pipecat ##### Runner - [Development Runner](https://docs.pipecat.ai/api-reference/server/utilities/runner/guide.md): Unified runner for building voice AI bots with Daily, WebRTC, and telephony transports - [Transport Utilities](https://docs.pipecat.ai/api-reference/server/utilities/runner/transport-utils.md): Configuration and helper utilities for Daily, LiveKit, telephony, and WebRTC transports ##### Service Utilities - [DailyRESTHelper](https://docs.pipecat.ai/api-reference/server/utilities/daily/rest-helper.md): Classes and methods for interacting with the Daily API to manage rooms and tokens ##### Service Switchers - [ServiceSwitcher](https://docs.pipecat.ai/api-reference/server/utilities/service-switchers/service-switcher.md): Dynamically switch between different service instances at runtime using configurable strategies - [LLMSwitcher](https://docs.pipecat.ai/api-reference/server/utilities/service-switchers/llm-switcher.md): Dynamically switch between different LLM services with support for ad-hoc inference and unified function registration ##### Smart Turn Detection - [Smart Turn Overview](https://docs.pipecat.ai/api-reference/server/utilities/turn-detection/smart-turn-overview.md): Advanced conversational turn detection powered by the smart-turn model - [Krisp VIVA Turn](https://docs.pipecat.ai/api-reference/server/utilities/turn-detection/krisp-viva-turn.md): Turn detection with the Krisp VIVA SDK in Pipecat as an alternative to VAD-based end-of-turn detection. ##### Telephony - [DTMFAggregator](https://docs.pipecat.ai/api-reference/server/utilities/dtmf-aggregator.md): Aggregates DTMF (phone keypad) input into meaningful sequences for LLM processing ##### Text Aggregators and Filters - [Text Aggregators and Filters](https://docs.pipecat.ai/api-reference/server/utilities/text/overview.md): An overview of text aggregators and filters available in the server utilities - [MarkdownTextFilter](https://docs.pipecat.ai/api-reference/server/utilities/text/markdown-text-filter.md): Converts Markdown-formatted text to TTS-friendly plain text while preserving structure - [PatternPairAggregator](https://docs.pipecat.ai/api-reference/server/utilities/text/pattern-pair-aggregator.md): Text aggregator that identifies and processes content between pattern pairs in streaming text ##### User and Assistant Turn Management - [User Turn Strategies](https://docs.pipecat.ai/api-reference/server/utilities/turn-management/user-turn-strategies.md): Configure how Pipecat detects and manages user turns: turn start and stop strategies for conversational timing. - [User Mute Strategies](https://docs.pipecat.ai/api-reference/server/utilities/turn-management/user-mute-strategies.md): Control when Pipecat suppresses user input: mute strategies during bot speech and other pipeline operations. - [Filter Incomplete User Turns](https://docs.pipecat.ai/api-reference/server/utilities/turn-management/filter-incomplete-turns.md): Use LLM-based detection to suppress responses when users are cut off mid-thought - [Turn Events](https://docs.pipecat.ai/api-reference/server/utilities/turn-management/turn-events.md): Handle user and assistant turn lifecycle events for transcriptions and turn tracking - [Transcriptions](https://docs.pipecat.ai/api-reference/server/utilities/turn-management/transcriptions.md): Collect user and assistant transcripts in Pipecat with turn-management transcription events and aggregators. - [External Turn Management](https://docs.pipecat.ai/api-reference/server/utilities/turn-management/external-turn-management.md): Handle turn detection externally using UserTurnProcessor or external services #### Events - [Events Overview](https://docs.pipecat.ai/api-reference/server/events/overview.md): Monitor Pipecat lifecycle changes with the event system: frame processor and service events, and how to subscribe. - [FrameProcessor Events](https://docs.pipecat.ai/api-reference/server/events/frame-processor-events.md): Handle errors and monitor frame processing with events available on every processor - [Service Events](https://docs.pipecat.ai/api-reference/server/events/service-events.md): Handle connection lifecycle and service-specific events for STT, TTS, and LLM services #### RTVI - [RTVI (Real-Time Voice Interaction)](https://docs.pipecat.ai/api-reference/server/rtvi/introduction.md): Build real-time voice and multimodal applications with Pipecat’s RTVI protocol - [RTVIProcessor](https://docs.pipecat.ai/api-reference/server/rtvi/rtvi-processor.md): RTVIProcessor coordinates RTVI protocol communication between Pipecat pipelines and connected clients. - [RTVIObserver](https://docs.pipecat.ai/api-reference/server/rtvi/rtvi-observer.md): RTVIObserver converts Pipecat pipeline frames into RTVI protocol messages for connected clients. - [Google RTVI Observer](https://docs.pipecat.ai/api-reference/server/rtvi/google-rtvi-observer.md): GoogleRTVIObserver adds support for sending Google search responses to RTVI clients in Pipecat pipelines. #### Frames - [Frames](https://docs.pipecat.ai/api-reference/server/frames/overview.md): Frame categories, processing behavior, and common patterns for Pipecat pipelines - [Data Frames](https://docs.pipecat.ai/api-reference/server/frames/data-frames.md): Reference for DataFrame types: audio, image, text, transcription, and transport messages - [Control Frames](https://docs.pipecat.ai/api-reference/server/frames/control-frames.md): Reference for ControlFrame types: pipeline lifecycle, response boundaries, service settings, and runtime configuration - [System Frames](https://docs.pipecat.ai/api-reference/server/frames/system-frames.md): Reference for SystemFrame types: pipeline lifecycle, interruptions, speaking state, input, and diagnostics - [LLM Frames](https://docs.pipecat.ai/api-reference/server/frames/llm-frames.md): LLM frame reference for Pipecat: context frames and the function calling helper dataclasses passed through pipelines. #### Workers - [BaseWorker](https://docs.pipecat.ai/api-reference/server/workers/base-worker.md): BaseWorker is the core agent class in Pipecat's multi-agent framework: lifecycle, bus access, and messaging. - [LLMWorker](https://docs.pipecat.ai/api-reference/server/workers/llm-worker.md): LLMWorker is a Pipecat agent with a built-in LLM pipeline and automatic tool registration for multi-agent systems. - [LLMContextWorker](https://docs.pipecat.ai/api-reference/server/workers/llm-context-worker.md): LLMContextWorker extends LLMWorker with a built-in conversation context and aggregators for stateful agents. - [UIWorker](https://docs.pipecat.ai/api-reference/server/workers/ui-worker.md): LLM agent that observes and drives a client GUI over the RTVI UI channel - [WorkerRunner](https://docs.pipecat.ai/api-reference/server/workers/runner.md): WorkerRunner orchestrates the lifecycle of Pipecat multi-agent systems: starting, supervising, and stopping workers. - [Types](https://docs.pipecat.ai/api-reference/server/workers/types.md): Shared types for Pipecat workers and job coordination: JobStatus, job context enums, and data classes. - [Exceptions](https://docs.pipecat.ai/api-reference/server/workers/exceptions.md): Error types for Pipecat job coordination: JobError and the exceptions raised by JobContext-based single-agent jobs. #### Pipeline - [PipelineParams](https://docs.pipecat.ai/api-reference/server/pipeline/pipeline-params.md): PipelineParams configures Pipecat pipeline execution: audio rates, metrics, heartbeats, and interruption settings. - [PipelineWorker](https://docs.pipecat.ai/api-reference/server/pipeline/pipeline-worker.md): PipelineWorker manages Pipecat pipeline execution and lifecycle inside the multi-agent framework. - [Pipeline Idle Detection](https://docs.pipecat.ai/api-reference/server/pipeline/pipeline-idle-detection.md): Detect and handle idle Pipecat pipelines with no bot or user activity, and shut them down automatically. - [Pipeline Heartbeats](https://docs.pipecat.ai/api-reference/server/pipeline/heartbeats.md): Monitor Pipecat pipeline health with heartbeat frames that verify frames still flow end to end. - [ParallelPipeline](https://docs.pipecat.ai/api-reference/server/pipeline/parallel-pipeline.md): Run multiple pipeline branches in parallel, with synchronized inputs and outputs for complex flows #### Bus - [Worker Bus](https://docs.pipecat.ai/api-reference/server/bus/bus.md): WorkerBus is the abstract base for inter-agent and runner-agent pub/sub messaging in Pipecat's multi-agent framework. - [Bus Messages](https://docs.pipecat.ai/api-reference/server/bus/messages.md): Bus message types for Pipecat multi-agent communication: the pub/sub primitives, organized by priority level. - [Bus Serializers](https://docs.pipecat.ai/api-reference/server/bus/serializers.md): MessageSerializer is the abstract base for Pipecat bus message serialization used by network bus implementations. - [Bus Proxy Agents](https://docs.pipecat.ai/api-reference/server/bus/proxy.md): WebSocketProxyClient and the bus proxy API: forward Pipecat bus messages to remote agents for distributed deployments. ### Client SDKs #### JavaScript SDK - [JavaScript SDK Overview](https://docs.pipecat.ai/api-reference/client/js/overview.md): The Pipecat JavaScript client SDK: device and media stream management, bot connectivity, and events for web voice AI apps. - [PipecatClient Constructor](https://docs.pipecat.ai/api-reference/client/js/client-constructor.md): Construct a PipecatClient from @pipecat-ai/client-js: transport selection, callbacks, and configuration options. - [Client Methods](https://docs.pipecat.ai/api-reference/client/js/client-methods.md): PipecatClient methods for starting bots, managing sessions, sending messages, and controlling devices in the JavaScript SDK. - [Callbacks and Events](https://docs.pipecat.ai/api-reference/client/js/callbacks.md): Handle bot events and state changes in the Pipecat JavaScript client with PipecatClient callbacks and event listeners. - [Errors](https://docs.pipecat.ai/api-reference/client/js/errors.md): Error types thrown by the Pipecat JavaScript client: RTVIError and its subclasses, and how to catch and handle each. ##### Transports - [Transport Overview](https://docs.pipecat.ai/api-reference/client/js/transports/transport.md): The Transport base class in the Pipecat JavaScript SDK: device management, connectivity, and messaging every transport implements. - [Daily WebRTC Transport](https://docs.pipecat.ai/api-reference/client/js/transports/daily.md): DailyTransport for the Pipecat JavaScript SDK: WebRTC audio and video on Daily's infrastructure, wrapping a Daily-JS call client. - [Small WebRTC Transport](https://docs.pipecat.ai/api-reference/client/js/transports/small-webrtc.md): SmallWebRTCTransport for the JavaScript SDK: peer-to-peer WebRTC to a Pipecat bot with no third-party WebRTC provider. - [WebSocket Transport](https://docs.pipecat.ai/api-reference/client/js/transports/websocket.md): WebSocketTransport for the JavaScript SDK: a lightweight, purely WebSocket-based connection between clients and a Pipecat bot. - [Gemini Live WebSocket Transport](https://docs.pipecat.ai/api-reference/client/js/transports/gemini.md): GeminiLiveWebsocketTransport for the JavaScript SDK: connect directly to Google's Gemini Live service over WebSocket. - [OpenAI Realtime WebRTC Transport](https://docs.pipecat.ai/api-reference/client/js/transports/openai-webrtc.md): OpenAIRealTimeWebRTCTransport for the JavaScript SDK: connect directly to the OpenAI Realtime API over WebRTC. #### React SDK - [React SDK Overview](https://docs.pipecat.ai/api-reference/client/react/overview.md): The Pipecat React SDK wraps PipecatClient with React components and hooks for building voice and multimodal AI applications. - [Components](https://docs.pipecat.ai/api-reference/client/react/components.md): Ready-to-use React components from the Pipecat React SDK: PipecatClientProvider, audio and video rendering, and visualizers. - [Hooks](https://docs.pipecat.ai/api-reference/client/react/hooks.md): React hooks from the Pipecat React SDK for client access, media device management, and bot event handling, like usePipecatClient. #### React Native SDK - [React Native SDK Overview](https://docs.pipecat.ai/api-reference/client/react-native/overview.md): The Pipecat React Native SDK brings PipecatClient from the JavaScript SDK to React Native voice and multimodal apps. - [React Native SDK Reference](https://docs.pipecat.ai/api-reference/client/react-native/api-reference.md): API reference for the Pipecat React Native SDK, which layers React Native support over the Pipecat JavaScript SDK. ##### Transports - [Daily WebRTC Transport](https://docs.pipecat.ai/api-reference/client/react-native/transports/daily.md): DailyTransport for the Pipecat React Native SDK: WebRTC audio and video on Daily's infrastructure in native mobile apps. - [Small WebRTC Transport](https://docs.pipecat.ai/api-reference/client/react-native/transports/small-webrtc.md): SmallWebRTCTransport for the React Native SDK: peer-to-peer WebRTC to a Pipecat bot with no third-party WebRTC provider. #### iOS SDK - [iOS SDK Overview](https://docs.pipecat.ai/api-reference/client/ios/overview.md): The Pipecat iOS SDK: a Swift client library for building voice and multimodal AI applications on iOS. - [iOS SDK Reference](https://docs.pipecat.ai/api-reference/client/ios/api-reference.md): Complete Swift API reference for the Pipecat iOS client SDK, hosted at docs-ios.pipecat.ai. ##### Transports - [Daily WebRTC Transport](https://docs.pipecat.ai/api-reference/client/ios/transports/daily.md): The Daily transport for the Pipecat iOS SDK: real-time WebRTC audio and video in Swift apps on Daily's infrastructure. - [Gemini Live WebSocket Transport](https://docs.pipecat.ai/api-reference/client/ios/transports/gemini-websocket.md): The Gemini Live WebSocket transport for the Pipecat iOS SDK: connect Swift apps directly to Google's Gemini Live service. - [OpenAI Realtime WebRTC Transport](https://docs.pipecat.ai/api-reference/client/ios/transports/openai-webrtc.md): OpenAIRealTimeWebRTCTransport for the Pipecat iOS SDK: connect Swift apps directly to the OpenAI Realtime API over WebRTC. - [Small WebRTC Transport](https://docs.pipecat.ai/api-reference/client/ios/transports/small-webrtc.md): SmallWebRTCTransport for the iOS SDK: peer-to-peer WebRTC between Swift clients and a Pipecat bot, no third-party provider. #### Android SDK - [Android SDK Overview](https://docs.pipecat.ai/api-reference/client/android/overview.md): The Pipecat Android SDK: a Kotlin client library for building voice and multimodal AI applications on Android. - [Android SDK Reference](https://docs.pipecat.ai/api-reference/client/android/api-reference.md): Complete Kotlin API reference for the Pipecat Android client SDK, hosted at docs-android.rtvi.ai. ##### Transports - [Daily WebRTC Transport](https://docs.pipecat.ai/api-reference/client/android/transports/daily.md): The Daily transport for the Pipecat Android SDK: real-time WebRTC audio and video in Kotlin apps on Daily's infrastructure. - [Gemini Live WebSocket Transport](https://docs.pipecat.ai/api-reference/client/android/transports/gemini-websocket.md): The Gemini Live WebSocket transport for the Pipecat Android SDK: connect Kotlin apps directly to Google's Gemini Live service. - [OpenAI Realtime WebRTC Transport](https://docs.pipecat.ai/api-reference/client/android/transports/openai-webrtc.md): The OpenAI Realtime WebRTC transport for the Pipecat Android SDK: connect Kotlin apps directly to the OpenAI Realtime service. - [Small WebRTC Transport](https://docs.pipecat.ai/api-reference/client/android/transports/small-webrtc.md): The Small WebRTC transport for the Android SDK: direct peer-to-peer WebRTC between Kotlin clients and a Pipecat bot. #### C++ SDK - [C++ SDK Overview](https://docs.pipecat.ai/api-reference/client/cpp/overview.md): The Pipecat C++ SDK: a native client library for voice and multimodal AI applications on Linux, macOS, and Windows. - [Daily WebRTC Transport](https://docs.pipecat.ai/api-reference/client/cpp/transport.md): The Daily transport for the Pipecat C++ SDK: real-time WebRTC audio and video in native applications on Daily's infrastructure. ### Pipecat Flows - [Pipecat Flows API Overview](https://docs.pipecat.ai/api-reference/pipecat-flows/overview.md): Reference documentation for Pipecat Flows: FlowManager, node configuration, actions, and the flows type system. - [FlowManager](https://docs.pipecat.ai/api-reference/pipecat-flows/flow-manager.md): FlowManager orchestrates Pipecat Flows conversations: node transitions, function registration, and flow state. - [Types](https://docs.pipecat.ai/api-reference/pipecat-flows/types.md): Type definitions and configuration schemas for Pipecat Flows: NodeConfig, task messages, and flow configuration structures. - [Exceptions](https://docs.pipecat.ai/api-reference/pipecat-flows/exceptions.md): The Pipecat Flows exception hierarchy: FlowError and subclasses raised during flow execution, and how to handle each. ### Pipecat Cloud #### REST Reference - [Pipecat Cloud REST API](https://docs.pipecat.ai/api-reference/pipecat-cloud/rest-reference/overview.md): Base URLs, authentication, error handling, and the OpenAPI specification for the Pipecat Cloud REST API. ##### Agents - [Create an Agent](https://docs.pipecat.ai/api-reference/pipecat-cloud/rest-reference/endpoint/agent-create.md): POST /agents creates a Pipecat Cloud agent from a container image with deployment configuration settings. - [Update an Agent](https://docs.pipecat.ai/api-reference/pipecat-cloud/rest-reference/endpoint/agent-update.md): POST /agents/{agentName} updates an existing Pipecat Cloud agent's configuration or deploys a new version of its image. - [Start an Agent Session](https://docs.pipecat.ai/api-reference/pipecat-cloud/rest-reference/endpoint/start.md): POST /{agentName}/start starts a new session with a deployed Pipecat Cloud agent. - [Stop an Agent Session](https://docs.pipecat.ai/api-reference/pipecat-cloud/rest-reference/endpoint/stop.md): DELETE /agents/{agentName}/sessions/{sessionId} stops a running Pipecat Cloud agent session and cleans up its resources. - [Session API](https://docs.pipecat.ai/api-reference/pipecat-cloud/rest-reference/endpoint/session-proxy.md): Proxy HTTP requests to endpoints defined in your running Pipecat Cloud bot: GET, POST, PUT, PATCH, DELETE, OPTIONS, and HEAD. - [Retrieve Agent Logs](https://docs.pipecat.ai/api-reference/pipecat-cloud/rest-reference/endpoint/agent-get-logs.md): GET /agents/{agentName}/logs returns execution logs for a Pipecat Cloud agent, with filtering and pagination options. - [List All Sessions](https://docs.pipecat.ai/api-reference/pipecat-cloud/rest-reference/endpoint/agent-get-sessions.md): GET /agents/{agentName}/sessions lists sessions for a Pipecat Cloud agent with filtering and pagination options. - [Get Session Details](https://docs.pipecat.ai/api-reference/pipecat-cloud/rest-reference/endpoint/agent-get-session.md): GET /agents/{agentName}/sessions/{sessionId} returns details for one agent session, including resource metrics. - [List All Agents](https://docs.pipecat.ai/api-reference/pipecat-cloud/rest-reference/endpoint/agent-list-all.md): GET /agents lists all Pipecat Cloud agents in your organization with their status and configuration. - [Get Agent Details](https://docs.pipecat.ai/api-reference/pipecat-cloud/rest-reference/endpoint/agent-list-one.md): GET /agents/{agentName} returns details for one Pipecat Cloud agent, including its deployment status and configuration. - [Delete an Agent](https://docs.pipecat.ai/api-reference/pipecat-cloud/rest-reference/endpoint/agent-delete.md): DELETE /agents/{agentName} permanently deletes a Pipecat Cloud agent and its associated resources. ##### Secrets - [Create or Update Secrets](https://docs.pipecat.ai/api-reference/pipecat-cloud/rest-reference/endpoint/secret-create-update.md): PUT /secrets/{setName} creates or updates a Pipecat Cloud secret set and its key-value pairs. - [List All Secret Sets](https://docs.pipecat.ai/api-reference/pipecat-cloud/rest-reference/endpoint/secret-list-all.md): GET /secrets lists all secret sets in your Pipecat Cloud organization with their names and metadata. - [Get Secret Set Details](https://docs.pipecat.ai/api-reference/pipecat-cloud/rest-reference/endpoint/secret-list-one.md): GET /secrets/{setName} returns the key-value pairs for a specific Pipecat Cloud secret set. - [Delete Entire Secret Set](https://docs.pipecat.ai/api-reference/pipecat-cloud/rest-reference/endpoint/secret-delete-secret-set.md): DELETE /secrets/{setName} deletes an entire Pipecat Cloud secret set and all the secrets it contains. - [Delete Specific Secret From Set](https://docs.pipecat.ai/api-reference/pipecat-cloud/rest-reference/endpoint/secret-delete-specific-secret.md): DELETE /secrets/{setName}/{secretKey} removes a single secret from a Pipecat Cloud secret set by its key. ##### Builds - [Get Upload URL](https://docs.pipecat.ai/api-reference/pipecat-cloud/rest-reference/endpoint/build-upload-url.md): POST /builds/upload-url returns a pre-signed URL for uploading your Docker build context to Pipecat Cloud. - [Create Build](https://docs.pipecat.ai/api-reference/pipecat-cloud/rest-reference/endpoint/build-create.md): POST /builds creates a new Pipecat Cloud build, or returns a cached build for your uploaded context. - [List Builds](https://docs.pipecat.ai/api-reference/pipecat-cloud/rest-reference/endpoint/build-list.md): GET /builds lists all Pipecat Cloud builds for your organization, with optional filters. - [Get Build](https://docs.pipecat.ai/api-reference/pipecat-cloud/rest-reference/endpoint/build-get.md): GET /builds/{buildId} returns the current status and details of a specific Pipecat Cloud build. - [Get Build Logs](https://docs.pipecat.ai/api-reference/pipecat-cloud/rest-reference/endpoint/build-get-logs.md): GET /builds/{buildId}/logs returns the build logs for a specific Pipecat Cloud build. ##### Regions - [List Regions](https://docs.pipecat.ai/api-reference/pipecat-cloud/rest-reference/endpoint/regions-list.md): GET /regions lists all available Pipecat Cloud regions for deploying agents and storing secrets. ##### Organization - [Get Organization Properties](https://docs.pipecat.ai/api-reference/pipecat-cloud/rest-reference/endpoint/properties-get.md): GET /properties returns the current values of configurable properties for your Pipecat Cloud organization. - [Update Organization Properties](https://docs.pipecat.ai/api-reference/pipecat-cloud/rest-reference/endpoint/properties-update.md): PATCH /properties updates configurable properties for your Pipecat Cloud organization, such as organization defaults. #### SDK Reference - [Pipecat Cloud SDK Overview](https://docs.pipecat.ai/api-reference/pipecat-cloud/sdk-reference/overview.md): The Pipecat Cloud Python SDK (pipecatcloud): programmatic agent deployment and session management from Python. - [Session Management](https://docs.pipecat.ai/api-reference/pipecat-cloud/sdk-reference/sessions.md): Start and manage Pipecat Cloud agent sessions programmatically with the pipecatcloud Python SDK's session APIs. - [Session Arguments](https://docs.pipecat.ai/api-reference/pipecat-cloud/sdk-reference/session-arguments.md): Understand the session arguments Pipecat Cloud passes to your bot entry point at session start, from the Python SDK. - [Error Handling](https://docs.pipecat.ai/api-reference/pipecat-cloud/sdk-reference/exceptions.md): Handle errors from the Pipecat Cloud Python SDK: exception types raised by deployment and session operations, and how to catch them. - [Examples](https://docs.pipecat.ai/api-reference/pipecat-cloud/sdk-reference/examples.md): Common Pipecat Cloud Python SDK patterns: starting agent sessions, handling session arguments, and managing deployments. ### CLI - [CLI Overview](https://docs.pipecat.ai/api-reference/cli/overview.md): The Pipecat CLI scaffolds new projects, runs behavioral evals, and deploys agents to Pipecat Cloud from your terminal. #### Commands - [pipecat init](https://docs.pipecat.ai/api-reference/cli/init.md): pipecat init scaffolds a new Pipecat app from the CLI: the single entry point for starting a project with Pipecat. - [pipecat eval](https://docs.pipecat.ai/api-reference/cli/eval.md): pipecat eval runs scenario-based behavioral evals against a running Pipecat agent, individually or as a suite from the CLI. - [pipecat context-hub](https://docs.pipecat.ai/api-reference/cli/context-hub.md): Query the Pipecat Context Hub from the CLI and register it as an MCP server for coding agents with pipecat context-hub install. ##### cloud - [pipecat cloud agent](https://docs.pipecat.ai/api-reference/cli/cloud/agent.md): pipecat cloud agent subcommands manage deployed Pipecat Cloud agents: view status and control agent lifecycle from the CLI. - [pipecat cloud auth](https://docs.pipecat.ai/api-reference/cli/cloud/auth.md): pipecat cloud auth manages authentication with Pipecat Cloud: login, logout, and checking your current credentials. - [pipecat cloud build](https://docs.pipecat.ai/api-reference/cli/cloud/build.md): pipecat cloud build manages Pipecat Cloud builds: view build logs, check build status, and inspect builds from the CLI. - [pipecat cloud docker](https://docs.pipecat.ai/api-reference/cli/cloud/docker.md): pipecat cloud docker builds, tags, and pushes agent Docker images to container registries for Pipecat Cloud deployments. - [pipecat cloud deploy](https://docs.pipecat.ai/api-reference/cli/cloud/deploy.md): pipecat cloud deploy creates or updates a Pipecat Cloud agent deployment, building a deployment manifest from your options. - [pipecat cloud organizations](https://docs.pipecat.ai/api-reference/cli/cloud/organizations.md): pipecat cloud organizations manages Pipecat Cloud organizations and API keys: list, select, and manage from the CLI. - [pipecat cloud secrets](https://docs.pipecat.ai/api-reference/cli/cloud/secrets.md): pipecat cloud secrets manages secret sets for agent deployments: create, update, list, and delete sensitive configuration. - [pipecat cloud regions](https://docs.pipecat.ai/api-reference/cli/cloud/regions.md): pipecat cloud regions lists the regions available for deploying agents and storing secrets in Pipecat Cloud. ### Pipecat Context Hub - [Pipecat Context Hub](https://docs.pipecat.ai/api-reference/context-hub.md): A local index of Pipecat docs, examples, and API source for AI coding tools — available as a CLI and an MCP server. ## Optional - [Pipecat Events](https://pipecat.ai/events) - [Community](https://discord.gg/pipecat) - [GitHub](https://github.com/pipecat-ai/pipecat) - [Changelog](https://github.com/pipecat-ai/pipecat/blob/main/CHANGELOG.md)