Daily

Build and deploy flexible real-time voice and multimodal AI agents across providers, transports, and platforms.

Deploy and scale production-ready voice and multimodal AI agents with integrated infrastructure, services, and operational tools.

Build scalable, cross-platform video and audio experiences with tools for real-time calling, streaming, recording, and analytics.

Build immersive, configurable audio experiences that scale to large audiences, with tools for recording, streaming, room management, and cross-platform development.

Add polished, reliable video calls to apps and websites quickly, with built-in collaboration features and support for mobile and large-scale experiences.

Reach large audiences with interactive, customizable live video, while combining real-time calls, broadcasts, and recordings through one platform.

Create flexible, branded recordings from realtime sessions, with options for custom layouts, separate tracks, and cloud or local capture.

Generate clinical notes through telehealth developer APIs to support clinical documentation workflows.

Pipecat

Pipecat is an open-source, vendor-neutral orchestration framework and ecosystem for building real-time voice and multimodal conversational AI agents. It includes server-side orchestration, cross-platform client SDKs, and components for connecting AI services and handling real-time interactions. Developers can use more than 40 AI services and APIs, WebRTC or WebSocket transports, and deployment environments including on-device and local/cloud hybrid applications.

Category: Conversational AI Platforms

• Build Real-Time AI Agent Pipelines
  Coordinate conversational AI agent components in a framework designed for real-time voice and multimodal applications.
• Build With Open-Source Components
  Use an open-source orchestration framework and ecosystem components, including SDKs, messaging and events standards, transport, and context engineering tools.
• Choose Providers And Transports
  Select or change AI services and transport options without being restricted to one vendor or transport.
• Connect 40+ AI Services
  Integrate a range of AI models and services, including LLM, STT, TTS, and multimodal capabilities.
• Manage Turns And Interruptions
  Support conversational flow with turn detection and interruption handling, including a native-audio turn detection model.
• Build Cross-Platform Client Apps
  Develop web and mobile client applications using SDKs with shared event abstractions and device-management capabilities.
• Connect Over WebRTC Or WebSockets
  Choose supported real-time connection methods and use SmallWebRTC for self-hosted or local development.
• Structure Agent Context
  Manage conversational context using framework capabilities and the Pipecat Flows context engineering library.
• Call Tools Asynchronously
  Use asynchronous tool calling within conversational agent pipelines.
• Coordinate Parallel Inference
  Coordinate parallel inference loops as part of real-time AI agent orchestration.
• Standardize Agent Support Interfaces
  Use standardized interfaces for observability, memory, and guardrails across agent pipelines.
• Build On-Device And Hybrid Apps
  Develop on-device or local/cloud hybrid applications across desktop and mobile platforms.

Pipecat Cloud

Pipecat Cloud is a managed platform for deploying Pipecat voice and multimodal AI agents on Daily infrastructure. It provides containerized deployment, multi-region autoscaling, observability, and integrated transport, telephony, noise cancellation, and recording services. Developers can connect agents to more than 80 AI service providers, use integrated services with bundled billing, or bring their own providers. The platform also includes enterprise security and compliance capabilities.

Category: Conversational AI Platforms

• Deploy Agents With Automatic Scaling
  Run containerized agents on multi-region infrastructure that scales automatically and supports unlimited concurrency.
• Connect To 80+ AI Providers
  Choose from a range of AI services or connect providers directly.
• Choose A Transport For Your Agents
  Use Daily WebRTC for voice or multimodal sessions, or connect a WebSocket provider.
• Support Enterprise Security Requirements
  Use stated HIPAA and GDPR compliance capabilities, enterprise-grade security and privacy, and options such as SSO and custom data residency.
• Connect Agents To Phone Calls
  Support phone-based voice agents with SIP and PSTN dial-in and dial-out, or use Twilio telephony.
• Monitor Deployed Agents
  Track deployed agents using built-in monitoring, telemetry, observability, and dashboards.
• Store Recordings And Transcriptions
  Keep recordings and session transcriptions in persistent storage.
• Reduce Noise And False Interruptions
  Reduce background noise and voices that can trigger false interruptions in agent conversations.
• Simplify Billing For Integrated Services
  Use bundled billing for integrated services, with discounted pricing and higher rate limits; bundled inference billing is also available to enterprise customers.
• Run Function Calls Server-Side
  Use server-side function calling in a Pipecat agent deployment.
• Integrate Simulation And Evaluation Tools
  Connect deployed agents with simulation and evaluation tools through platform integrations.

Daily Video SDK

Daily Video SDK provides cross-platform SDKs and WebRTC infrastructure for building real-time video and audio applications. Developers can build for JavaScript, React Native, iOS, Android, Python, Flutter, and other platforms, with APIs for live streaming, recording, media tracks, and call management. Its global infrastructure supports low-latency calls and up to 100,000 active participants, while analytics, security, and compliance capabilities help teams operate and monitor their applications.

Category: Communications Platform as a Service

• Build Across Major Platforms
  Choose SDKs for web, mobile, desktop, and server applications, or explore custom builds for other use cases.
• Connect Participants At Scale
  Build real-time applications on global infrastructure supporting up to 100,000 active participants in a session.
• Adapt Video To Network Conditions
  Use bandwidth management, simulcast, and codec controls to support video quality across varied networks and devices.
• Deliver Efficient iOS Video
  Offer HEVC video on supported iOS devices while retaining H.264 compatibility for devices that do not support HEVC.
• Customize Calls And Media
  Manage media tracks and build call experiences tailored to application requirements.
• Stream Calls To More Viewers
  Extend real-time experiences to third-party streaming platforms.
• Create Custom Recordings
  Record calls with custom layouts and graphics, and choose composite or individual media tracks and supported formats.
• Support Secure Applications
  Use encrypted calls and infrastructure and compliance capabilities to support enterprise security requirements.
• Monitor Calls And Performance
  Review logs and metrics in the dashboard, use the REST API, or connect data to enterprise BI platforms.
• Share Updates Across Platforms
  Build on a common core designed to support consistent experiences and cross-platform SDK updates.
• Build AI-Enabled Workflows
  Connect video applications with transcription, recording, and AI integrations for supported workflows.
• Join Calls By Phone
  Allow participants to join calls by dialing a phone number.

Daily Audio SDK

Daily Audio SDK provides cross-platform APIs and global infrastructure for building real-time audio applications, including broadcast, social, and metaverse experiences. Developers can support up to 100,000 active participants in one space, configure stereo audio quality, record sessions, and multicast to third-party platforms using one API. The SDK also includes room controls, analytics, audio processing, and tools for spatial audio and AI integrations.

Category: Communications Platform as a Service

• Bring Together Up To 100,000 Participants
  Host large audio experiences in a single space with up to 100,000 active participants.
• Deliver Low-Latency Audio Globally
  Run audio experiences on global infrastructure with a reported median first-hop latency under 13 milliseconds.
• Configure Stereo Audio For Your Needs
  Tune stereo bitrates and select tested profiles for different use cases and low-bandwidth conditions.
• Reduce Noise And Manage Audio
  Use audio processing and noise cancellation to manage sound in audio experiences.
• Build Audio Apps Across Platforms
  Develop audio experiences using cross-platform SDKs, including native mobile SDKs with a shared Rust core.
• Record Audio Locally Or In The Cloud
  Choose local or cloud recording, or send raw audio tracks directly to an S3 bucket.
• Stream Rooms Over HLS Or RTMP
  Broadcast audio rooms over HLS or RTMP.
• Create Proximity-Based 3D Audio
  Build audio for 3D worlds by selecting tracks and adjusting their volume based on proximity.
• Manage Audio Tracks Per Client
  Control which audio tracks clients receive and manage their volumes independently.
• Protect Audio With Encryption
  Protect media transport with AES-256 and calls with TLS encryption; enable end-to-end encryption in peer-to-peer mode.
• Customize Raw Audio Processing
  Work directly with raw tracks to implement processing before or after a session.
• Connect Audio Apps To AI Services
  Integrate hosted AI services, LLMs, or proprietary machine-learning infrastructure, with transcription and noise cancellation among the standard AI features.
• Manage Participants And Room Settings
  Set meeting limits, eject listeners, promote listeners to moderators, and let participants raise their hands.
• Monitor Audio App Performance
  Review logs and metrics in the dashboard, use the REST API, or leverage BI integrations.

Daily Prebuilt

Daily Prebuilt is an embeddable video call interface that developers can add to an app or website in minutes with a few lines of code. It provides responsive, no-download calls with built-in participant features, configurable layouts, and tools for recording, analytics, and security.

• Embed Calls In Minutes
  Add a ready-made calling interface with minimal code and no participant software download.
• Scale Calls To 100,000 Participants
  Build experiences for small meetings or large audiences with real-time participant interaction.
• Optimize Calls Across Networks
  Use automatic bandwidth management and video infrastructure to support call quality across networks and devices.
• Engage Participants In Calls
  Support communication and collaboration with chat, reactions, hand raising, breakout rooms, and custom interactions.
• Support Calls Across Devices
  Provide a calling interface designed to adapt to different devices and screen sizes.
• Protect Calls And Data
  Use stated SOC 2, GDPR, and HIPAA support, with options including end-to-end encryption, peer-to-peer routing, and data geofencing.
• Share Screens In HD
  Let participants share screens during calls, including dual screen sharing.
• Configure Call Layouts
  Use active-speaker views, configurable grids, and pinning to shape participant displays.
• Reduce Noise And Background Distractions
  Use noise reduction and background effects to support clearer audio and less distracting calls.
• Prepare And Recover From Call Issues
  Let users check their camera and microphone before calls and help them recover from common permission and network problems.
• Record Calls In 1080P
  Capture calls with high-quality 1080p cloud recording and available recording options.
• Localize The User Experience
  Offer the interface in 15 languages.
• Control Participant Permissions
  Use roles and permissions to shape workflows and control access to camera, microphone, and screen sharing.
• Monitor Call Performance
  Use analytics, metrics, logging, and monitoring tools to review call performance and support users.
• Customize The Call Interface
  Configure the interface and apply color themes to create a call experience suited to an app.
• Access Enterprise Support Plans
  Choose enterprise SLAs and support plans for the organization's use case and scale.

Daily Live Streaming

Daily Live Streaming gives developers a unified API for combining real-time WebRTC experiences with HLS or RTMP broadcasts. It supports interactive sessions with up to 100,000 active participants and broadcasts to millions of viewers. Daily's Video Component System enables custom layouts and event-driven overlays for live and streamed content. The platform supports multi-platform distribution, cloud recording, multiple output formats, and live audience participation, including presenters, co-hosts, polls, reactions, and other interactions.

• Host Up To 100,000 Participants In Real Time
  Bring large audiences into interactive video experiences with real-time video and low latency.
• Bring Audiences Into The Experience
  Let viewers participate in real time by joining the stage, responding with reactions, taking part in polls, or engaging in activities such as auctions and bidding.
• Broadcast With HLS Or RTMP
  Turn real-time video or audio experiences into broadcasts using HLS or RTMP.
• Bring Presenters And Co-Hosts Onstage
  Support unlimited presenters and co-hosts and bring audience participants onto the stage.
• Deliver Real-Time Video Globally
  Use globally distributed infrastructure for low-latency video delivery across real-time sessions and streaming experiences.
• Reach Viewers Across Multiple Platforms
  Distribute live output to multiple supported destinations at once, including social platforms and custom media servers.
• Customize Layouts And Add Overlays
  Shape live and streamed content with branded layouts, dynamic graphics, and event-driven visual elements using the Video Component System.
• Combine Calls And Broadcasts
  Connect interactive real-time video experiences with broadcasts for larger audiences through a unified streaming workflow.
• Tailor Layouts To Each Device
  Design compositions for portrait, landscape, and other device and platform needs.
• Record Customizable Live Experiences
  Extend live content into recordings with customized layouts and dynamic elements that match the live experience.
• Add Streaming With One API
  Integrate live streaming into web or native applications through Daily's API.
• Build In-App Production Studios
  Enable creators to produce interactive videos within an application without expensive hardware or complicated desktop production software.

Daily Recording

Daily Recording is a programmable toolkit for creating customizable recordings and live streams from realtime sessions. It supports HD cloud recordings in MP4 or HLS, individual audio and video tracks saved to cloud storage, and local recording on client devices. The Video Component System provides custom layouts, animated graphics, and event-driven overlays. Developers can control recording and layout options through client-side code or the REST API, choose Daily Cloud or their own Amazon S3 bucket, and configure multiple formats, outputs, and orientations.

• Save HD Recordings To Cloud Or S3
  Capture video and audio in MP4 or HLS and store it in Daily Cloud or an Amazon S3 bucket.
• Customize Layouts And Animated Graphics
  Build branded recordings with custom layouts, animated graphics, event-driven overlays, and rendering choices adapted to platform and device.
• Capture Separate Participant Media Tracks
  Save raw audio and video streams as individual tracks for post-processing, editing, compliance, and AI use cases.
• Render Smooth, Consistent Composites
  Use a shared compositor for cloud recording and raw-track post-processing, with layout transitions that move video layers and change opacity at 30 fps.
• Generate Multiple Formats And Orientations
  Create multiple recording configurations, including portrait and landscape outputs and composite recordings alongside individual tracks.
• Control Recording Through APIs
  Configure recording and layout behavior in code or through the REST API, including supported changes during a session.
• Reconstruct Meetings With Event Data
  Use event JSON alongside raw media tracks to provide meeting context, including participant activity, active speakers, transcripts, and application messages.
• Prepare Gapless Audio For Processing
  Create uniform WAV or AAC files with silence mixed in and no data gaps for use in audio-processing applications.
• Record In Configurable 1080p Formats
  Create 1080p recordings and configure MP4 bitrates or HLS bitrate ladders.
• Choose Storage For Compliance Needs
  Select cloud regions or a customer-owned S3 bucket for recording storage according to compliance requirements, including HIPAA and GDPR.
• Record Tracks On Client Devices
  Capture individual audio and video tracks locally on client devices.

AI-powered Clinical Notes API

An API for generating clinical notes, integrated into Daily's telehealth developer APIs and powered by AI tools.

• Generate Clinical Notes
  Use AI-powered clinical note generation integrated into Daily's telehealth developer APIs.