daily - Build real-time voice and multimodal AI pipelines
Provides guidance for building real-time voice and multimodal AI applications with pipelines, transports, AI services, tools, and deployment options.
Tags
Updated: 2026-09-29Capabilities
Typical Inputs
Typical Outputs
What this skill does
- Build real-time voice conversations
- Process multimodal audio and video
- Integrate AI service providers
- Connect external functions and APIs
- Manage conversation context
- Configure transport connections
- Detect turns and interruptions
- Create custom frame processors
- Build structured conversation flows
- Monitor latency and usage
- Integrate frontend client SDKs
- Deploy and scale applications
Inputs
- User audio
- User video
- User text
- AI service configurations
- Transport configurations
- Conversation context
- Tool definitions
- External API credentials
- Session parameters
Outputs
- Audio responses
- Video responses
- Text responses
- Speech transcriptions
- Function call results
- Conversation context updates
- Session lifecycle events
- Latency and usage metrics
- Distributed traces