llms.txt

# 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 Enterprise**: The same Pipecat Cloud control plane, with agents
  running in a Kubernetes cluster you operate. See the
  [Enterprise overview](https://docs.pipecat.ai/pipecat-cloud/enterprise/overview.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, the client SDKs, and Pipecat Cloud managed 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 Cloud and Enterprise Hosting](https://docs.pipecat.ai/overview/cloud.md): Hosted infrastructure for Pipecat agents: Pipecat Cloud runs them in Daily-hosted regions, Pipecat Enterprise in your own VPC.

## 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.
- [Migrating from pipecat-ai-flows](https://docs.pipecat.ai/pipecat/migration/flows.md): Move a Pipecat Flows app off the standalone pipecat-ai-flows package: new pipecat.flows imports and the deprecated APIs that were removed.

### 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
- [Error Handling](https://docs.pipecat.ai/pipecat/fundamentals/error-handling.md): Handle service failures in Pipecat with is_usable, ErrorCategory, and ProcessorUnusablePolicy, and recover a failed service.
- [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.

### Flows

- [Pipecat Flows](https://docs.pipecat.ai/pipecat/flows/introduction.md): Structure a conversation as a graph of nodes with Pipecat Flows: declarative flow configs, programmatic nodes, and transitions.
- [Pipecat Flows Quickstart](https://docs.pipecat.ai/pipecat/flows/quickstart.md): Build your first Pipecat Flows conversation: a two-node Hello World flow written as a flow config with one Python handler.
- [Flow Configs](https://docs.pipecat.ai/pipecat/flows/flow-configs.md): Write a Pipecat Flows conversation as data with FlowConfig: the document format, loading, validation, and joining a config to handlers.
- [Nodes & Messages](https://docs.pipecat.ai/pipecat/flows/nodes-and-messages.md): Configure Pipecat Flows conversation nodes: role and task messages, respond_immediately, and how nodes shape each step.
- [Functions](https://docs.pipecat.ai/pipecat/flows/functions.md): Define Pipecat Flows tools: node and edge functions, transition_only entries, branch tables, and the direct-function handler contract.
- [Actions](https://docs.pipecat.ai/pipecat/flows/actions.md): Run custom functionality at set points in a Pipecat Flows node: built-in tts_say, end_conversation, function, and custom actions.
- [Context Strategies](https://docs.pipecat.ai/pipecat/flows/context-strategies.md): Manage Pipecat Flows conversation context across node transitions with the append and reset strategies.
- [State Management](https://docs.pipecat.ai/pipecat/flows/state-management.md): Share data across Pipecat Flows nodes: FlowManager state, prompt placeholders, and global functions.
- [Pipecat Flows Examples](https://docs.pipecat.ai/pipecat/flows/examples.md): Pipecat Flows examples grouped by form: declarative flow configs with handlers, and flows written in Python where code is needed.

### Evals

- [Pipecat Evals](https://docs.pipecat.ai/pipecat/evals/overview.md): Test agent behavior with scripted conversations and simulated callers, run against the same agent you deploy.
- [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 evals against an existing agent: a scripted scenario, then a simulated caller with a goal.
- [Scripted Scenarios](https://docs.pipecat.ai/pipecat/evals/scripted-scenarios.md): The scripted scenario format: user turns, expected events, eval, text_contains and llm_marker assertions, function calls, and latency budgets.
- [Simulated Scenarios](https://docs.pipecat.ai/pipecat/evals/simulated-scenarios.md): The simulated scenario file format: a persona and goal an LLM plays against your agent, a success criterion, judged and measured metrics, and runs.
- [Scenario Configuration](https://docs.pipecat.ai/pipecat/evals/scenario-configuration.md): Scenario files for scripted and simulated scenarios: the scenarios: list, file defaults, user: and judge: blocks, factory:, and !include.
- [Eval Suites](https://docs.pipecat.ai/pipecat/evals/suites.md): Spawn agents and run many scripted and simulated scenarios concurrently from a single manifest with pipecat eval suite.
- [Using the Library](https://docs.pipecat.ai/pipecat/evals/library.md): Run scripted scenarios and simulations from Python with the pipecat.evals API: sessions, results, custom judges and persona LLMs, and suites.
- [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.

### Pipecat UI

- [Pipecat UI](https://docs.pipecat.ai/client/pipecat-ui.md): Customizable React components for voice AI interfaces, installed as source through the Pipecat UI shadcn registry.

## Pipecat Cloud & Enterprise

### Get Started

- [Pipecat Cloud & Pipecat Enterprise](https://docs.pipecat.ai/pipecat-cloud/introduction.md): Run Pipecat agents in production: Pipecat Cloud in Daily-hosted regions, or Pipecat Enterprise with a self-hosted region in your VPC.

### 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.
- [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.
- [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

#### Building & Deploying

- [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.
- [Deploy from GitHub](https://docs.pipecat.ai/pipecat-cloud/guides/deploy-from-github.md): Connect the Pipecat Cloud GitHub App to deploy agents from a repository branch, with deploy on push, manual deploys, and commit tracking.
- [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.

##### 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 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.
- [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 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.

#### Webhooks

- [Pipecat Cloud Webhooks](https://docs.pipecat.ai/pipecat-cloud/guides/webhooks.md): Register webhook endpoints, subscribe to Pipecat Cloud events, and verify signed webhook deliveries in TypeScript and Python.
- [Webhook Event Reference](https://docs.pipecat.ai/pipecat-cloud/guides/webhook-events.md): Every Pipecat Cloud webhook event: the delivery envelope plus payloads for build, deployment, service, spend, session, API key, and secret events.

#### 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.

### Pipecat Enterprise

- [Pipecat Enterprise](https://docs.pipecat.ai/pipecat-cloud/enterprise/overview.md): Run Pipecat Cloud agents in your own Kubernetes cluster with Pipecat Enterprise: what stays managed, what you operate, what leaves your cluster.
- [Pipecat Enterprise Architecture](https://docs.pipecat.ai/pipecat-cloud/enterprise/architecture.md): How a self-hosted region is split between the Pipecat Cloud control plane and your data plane: what each part does, sees, and connects to.
- [What Pipecat Cloud Can See](https://docs.pipecat.ai/pipecat-cloud/enterprise/what-daily-can-see.md): Exactly what data the Pipecat Cloud control plane reads from a self-hosted region, what it never reads, and how to turn each channel off.

#### Set up a region

- [Pipecat Enterprise Requirements](https://docs.pipecat.ai/pipecat-cloud/enterprise/requirements.md): Cluster, platform, network egress, and node requirements for a Pipecat Cloud self-hosted region, with sizing guidance for platform components.
- [Set Up a Pipecat Enterprise Region](https://docs.pipecat.ai/pipecat-cloud/enterprise/setup.md): Register a self-hosted region, mint an enrollment token and registry key, install the region package with Helm, and verify the region is active.
- [Pipecat Enterprise Configuration](https://docs.pipecat.ai/pipecat-cloud/enterprise/configuration.md): Region package Helm values reference: workloads namespace, external broker and log store, WebSocket exposure, replica counts, and pull secrets.
- [Pipecat Enterprise Secrets](https://docs.pipecat.ai/pipecat-cloud/enterprise/secrets.md): Reference Kubernetes Secrets in your cluster as Pipecat Cloud secret sets: env and image-pull secrets, readiness states, and what is recorded.
- [Pipecat Enterprise Agent Sizing](https://docs.pipecat.ai/pipecat-cloud/enterprise/sizing.md): Size agents in self-hosted regions with explicit CPU and memory or organization agent profiles, and target amd64 or arm64 nodes per deployment.
- [Pipecat Enterprise WebSockets](https://docs.pipecat.ai/pipecat-cloud/enterprise/websockets.md): Expose a self-hosted region's WebSocket endpoint for telephony and client sessions: Ingress or Gateway API, the public endpoint, token auth.

#### Operate

- [Pipecat Enterprise Logs](https://docs.pipecat.ai/pipecat-cloud/enterprise/logs.md): Agent logs and session metrics in self-hosted regions: enabling the optional log store, base image requirements, cgroup v2, and platform metrics.
- [Operating Pipecat Enterprise](https://docs.pipecat.ai/pipecat-cloud/enterprise/operations.md): Day-two operations for a self-hosted region: package upgrades, certificate lifecycle, high availability, rebuilding a cluster, and retiring a region.
- [Pipecat Enterprise Diagnostics](https://docs.pipecat.ai/pipecat-cloud/enterprise/diagnostics.md): Diagnose a self-hosted region with the doctor, region status and event feed, and the support bundle, and what to collect before contacting support.

#### Reference architectures

- [Pipecat Enterprise on AWS EKS](https://docs.pipecat.ai/pipecat-cloud/enterprise/reference/aws-eks.md): A reference architecture for a self-hosted region on Amazon EKS: VPC, node pools, managed broker and log store, WebSocket front door, monitoring.
- [Pipecat Enterprise on Oracle OKE](https://docs.pipecat.ai/pipecat-cloud/enterprise/reference/oci-oke.md): A reference architecture for a self-hosted region on Oracle Cloud's OKE: standalone VCN, Karpenter node pools, managed stores, Gateway API front door.

### 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.

### Support

- [Support & Professional Services](https://docs.pipecat.ai/pipecat-cloud/support.md): Support and professional services for Pipecat and Pipecat Cloud from Daily: architecture guidance, production readiness, and compliance.

## 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.
- [MCP Transport](https://docs.pipecat.ai/api-reference/server/services/transport/mcp.md): pipecat-mcp-transport serves a Pipecat bot as an MCP server, so any agent with an MCP client talks to it.
- [Media over QUIC Transport](https://docs.pipecat.ai/api-reference/server/services/transport/moq.md): Real-time audio over Media over QUIC with MOQTransport and MOQParams: serve mode, relay client mode, and TLS.
- [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): SingleClientWebsocketServerTransport 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): AssemblyAI STT services: streaming WebSocket API (AssemblyAISTTService) and segmented HTTP API (AssemblyAISyncSTTService) 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): Streaming STT with GeminiSTTService on the Gemini Live API and GoogleSTTService on Cloud Speech-to-Text V2: hints and 125+ languages.
- [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.
- [Meta Speech-to-Text](https://docs.pipecat.ai/api-reference/server/services/stt/meta.md): Streaming STT with MetaSTTService: Muse Voice realtime transcription with 25 language support and three segmentation modes.
- [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.
- [Oruk Speech-to-Text](https://docs.pipecat.ai/api-reference/server/services/stt/oruk.md): Oruk realtime speech and phrase estimates, plus optional local Orukeet CPU transcription for completed utterances.
- [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 and SarvamRealtimeSTTService stream real-time STT for Indian languages with VAD-based or server-side endpointing.
- [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 turn detection via Agent STT API.
- [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): BYOK-first LLM for Pipecat — route your own OpenAI key through Floe for per-call metering and spend caps, or go keyless.
- [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.
- [Bland Text-to-Speech](https://docs.pipecat.ai/api-reference/server/services/tts/bland.md): Realtime TTS with BlandTTSService (WebSocket streaming) and BlandHttpTTSService for complete-text 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 and Flux TTS with DeepgramTTSService, DeepgramHttpTTSService, DeepgramFluxTTSService, and SageMaker variants for hosted inference.
- [ElevenLabs Text-to-Speech](https://docs.pipecat.ai/api-reference/server/services/tts/elevenlabs.md): TTS with ElevenLabsTTSService (WebSocket), ElevenLabsHttpTTSService (HTTP), and ElevenLabsDialogueTTSService for Eleven v3.
- [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): BYOK-first text-to-speech for Pipecat — route your own OpenAI key through Floe for per-call metering and spend caps, or go keyless.
- [Gandr Text-to-Speech](https://docs.pipecat.ai/api-reference/server/services/tts/gandr.md): GandrTTSService streams speech over a persistent WebSocket, holding one connection warm across turns.
- [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 and configurable sample rates 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.
- [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.
- [Maya Text-to-Speech](https://docs.pipecat.ai/api-reference/server/services/tts/maya.md): Stream Indian-language TTS with MayaTTSService, or use MayaHttpTTSService for HTTP synthesis, in a Pipecat pipeline.
- [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.
- [SILMA Text-to-Speech](https://docs.pipecat.ai/api-reference/server/services/tts/silma.md): SilmaTTSService streams Arabic and English speech from SILMA AI's low-latency WebSocket 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.
- [SonexLabs Text-to-Speech](https://docs.pipecat.ai/api-reference/server/services/tts/sonex.md): SonexTTSService converts text to speech with SonexLabs' streaming TTS API.
- [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.
- [Speechify Text-to-Speech](https://docs.pipecat.ai/api-reference/server/services/tts/speechify.md): HTTP-based TTS with SpeechifyHttpTTSService: word-level timestamps via Server-Sent Events using simba-3.2 and simba-3.0 models.
- [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.
- [Vakyam AI Text-to-Speech](https://docs.pipecat.ai/api-reference/server/services/tts/vakyam.md): Add Tamil, Hindi, Telugu, and other Indian-language voices to a Pipecat pipeline with VakyamTTSService.
- [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.
- [Boson Higgs Realtime](https://docs.pipecat.ai/api-reference/server/services/s2s/boson.md): Real-time speech-to-speech service powered by Boson AI's Higgs Realtime API
- [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 GPT-Live](https://docs.pipecat.ai/api-reference/server/services/s2s/openai-live.md): OpenAILiveLLMService connects Pipecat to OpenAI's GPT-Live API for full-duplex speech-to-speech conversation with delegated tools.
- [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.
- [ThunderPhone](https://docs.pipecat.ai/api-reference/server/services/s2s/thunderphone.md): Speech-to-speech service that runs a ThunderPhone voice agent (speech recognition, LLM, voice, turn-taking, 47 languages and tools) inside a Pipecat pipeline
- [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.
- [MemorySync Memory](https://docs.pipecat.ai/api-reference/server/services/memory/memorysync.md): The pipecat-memorysync integration gives Pipecat voice agents long-term, cross-session memory backed by MemorySync — budgeted recall that never stalls a reply.
- [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.

##### Agents

- [OpenClaw Gateway](https://docs.pipecat.ai/api-reference/server/services/agents/openclaw.md): Drive an OpenClaw coding agent from a pipeline with OpenClawGatewayService: start, steer, and abort agent runs.

##### 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.
- [floe-guard](https://docs.pipecat.ai/api-reference/server/services/analytics/floe.md): A local budget guardrail for Pipecat voice agents — prices every leg (STT, LLM, TTS, telephony) and hard-stops the next turn before it crosses a spend ceiling. No account, no network.
- [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.
- [MLflow](https://docs.pipecat.ai/api-reference/server/services/analytics/mlflow.md): Trace and analyze Pipecat voice agent conversations with MLflow
- [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

- [Keenable Web Search](https://docs.pipecat.ai/api-reference/server/services/knowledge-retrieval/keenable.md): KeenableWebSearch provides low-latency web search and page reading tools via a hosted MCP server powered by Keenable AI.
- [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.
- [Hecttor](https://docs.pipecat.ai/api-reference/server/services/audio-filters/hecttor.md): Real-time speech enhancement and audio denoising using the Hecttor SDK
- [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.
- [pipecat-effects](https://docs.pipecat.ai/api-reference/server/services/audio-filters/pipecat-effects.md): Audio effects for the output of a pipecat bot
- [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.
- [Error Observer](https://docs.pipecat.ai/api-reference/server/utilities/observers/error-observer.md): ErrorObserver reports every error a pipeline raises through on_error as an ErrorEvent with category, processor, and usability info.
- [Function Call Observer](https://docs.pipecat.ai/api-reference/server/utilities/observers/function-call-observer.md): FunctionCallObserver reports each function call a conversation makes, from start to outcome, as FunctionCallEvent moments.
- [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.
- [Service Metrics Observer](https://docs.pipecat.ai/api-reference/server/utilities/observers/service-metrics-observer.md): Track service latency and usage metrics with ServiceMetricsObserver for STT, TTS, and LLM services in Pipecat
- [Speaking Observer](https://docs.pipecat.ai/api-reference/server/utilities/observers/speaking-observer.md): SpeakingObserver reports speaking lifecycle moments: user speech and turn events, bot speech, and interruptions with accurate timestamps.
- [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
- [Voice Formatter](https://docs.pipecat.ai/api-reference/server/utilities/text/voice-formatter.md): VoiceFormatter rewrites currency, dates, acronyms, and phone numbers into spoken form before TTS synthesis.

##### 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.
- [BaseUIWorker](https://docs.pipecat.ai/api-reference/server/workers/base-ui-worker.md): Worker that surfaces its jobs and job groups on the client UI as progress cards, without involving an LLM.
- [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.
- [LiveKit Transport](https://docs.pipecat.ai/api-reference/client/js/transports/livekit.md): LiveKitTransport for the Pipecat JavaScript SDK: WebRTC audio and video on LiveKit's infrastructure, wrapping a LiveKit client.
- [MoQ Transport](https://docs.pipecat.ai/api-reference/client/js/transports/moq.md): MoqTransport for the JavaScript SDK: Media over QUIC audio and RTVI messaging to a Pipecat bot, over WebTransport.
- [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.
- [FlowConfig](https://docs.pipecat.ai/api-reference/pipecat-flows/flow-config.md): FlowConfig reference for Pipecat Flows: the declarative flow document, its node and function models, loaders, and validation rules.
- [Flow](https://docs.pipecat.ai/api-reference/pipecat-flows/flow.md): Flow reference for Pipecat Flows: join a FlowConfig to your handlers, resolve tool references, and feed nodes to FlowManager.
- [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): start and manage agent sessions, type bot() entry points, and handle SDK errors.
- [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 an agent session, and writing bot() entry points for Daily and WebSocket.

### 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 scripted and simulated 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

- [Output Modes and Exit Codes](https://docs.pipecat.ai/api-reference/cli/cloud/output.md): Render pipecat cloud output as rich, plain, or json, read its exit codes, and run it non-interactively in CI.
- [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 github](https://docs.pipecat.ai/api-reference/cli/cloud/github.md): pipecat cloud github connects an organization to GitHub, inspects the App installation, and lists visible repositories and branches.
- [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 spend-limit](https://docs.pipecat.ai/api-reference/cli/cloud/spend-limit.md): pipecat cloud spend-limit sets, shows, and clears an organization's spending cap, blocking new sessions once the cap is reached.
- [pipecat cloud regions](https://docs.pipecat.ai/api-reference/cli/cloud/regions.md): pipecat cloud regions lists the regions available for deploying agents and manages the lifecycle of self-hosted regions.

### 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)