Real Time Video And Audio Communication AI Tool
How do I start building a voice AI agent with LiveKit?
Voice AI quickstart: Build a simple voice agent with Python or Node.js in less than 10 minutes. Try it out. There are voice agent starter apps to help you bring your agent to life through a web or mobile app, you can integrate with telephony, and you can deploy to LiveKit Cloud.
What is LiveKit's Agent Platform and what does it include?
- Open source framework to build and test agents
- Inference gateway to access TTS, LLM, and STT models
- Cloud platform for deploying and scaling agents
- Phone numbers and SIP integrations for telephony
- Full-stack observability for every agent session
How do I deploy to LiveKit Cloud and scale my agents in production?
Deploy to LiveKit Cloud and run your agents on LiveKit's global realtime infrastructure. This enables you to deploy, monitor, and scale your voice and video agents in production.
Can LiveKit connect to telephony for phone calls?
Yes. LiveKit provides phone numbers and SIP integrations for telephony, enabling your voice agent to make or take phone calls.
How does LiveKit support real-time voice AI applications with low latency?
LiveKit supports voice AI applications by providing tools to build real-time multimodal experiences with sub-100ms latency. It enables streaming of voice and video between models and users, with features like turn-taking cue detection, interruption handling, and telephony network connectivity to enhance natural interactions.
What security and compliance features does LiveKit provide?
LiveKit offers enterprise-grade security with end-to-end encryption and compliance with SOC 2 Type 2, GDPR, CCPA, and HIPAA. It supports SSO and role-based access control, and includes noise and echo cancellation, session recording, and moderation tools to maintain a secure, high-quality experience.
What is the pricing for LiveKit and what is the free tier?
- Real-time audio and video: $0.18 per GB
- Egress/Ingress: $0.02 per minute
- Recording and composition: $0.02 per minute
- Initial allocation of 50GB free monthly bandwidth
- Volume discounts:
- 0–50GB: Free
- 50–1,000GB: $0.18/GB
- 1,000–10,000GB: $0.16/GB
- 10,000–50,000GB: $0.14/GB
- 50,000–100,000GB: $0.12/GB
- 100,000GB+: Custom pricing
- Free agent session minutes: 1,000 per month
- No credit card required
What are the limitations of the Free Plan and quotas?
- Free plans: up to 100 concurrent connections
- Egress time limits: 3 hours for file outputs; 12 hours for HLS segments and streaming
- Subscription limits: up to 50 video tracks and 100 audio tracks per participant
- Free quotas: projects on the free plan cease functioning after exceeding 50GB data transfer or 60 minutes of transcoding
Where can I access documentation and starter resources?
Documentation is available on LiveKit. There are specific resources for getting started with voice AI (Voice AI quickstart), voice agent starter apps, telephony integration guides, and workflows for deploying to LiveKit Cloud.
What is the relationship between LiveKit's open-source framework and LiveKit Cloud?
LiveKit provides an open-source framework to build and test agents, plus an inference gateway to access TTS, LLM, and STT models. The Cloud platform is a managed service for deploying, scaling, and observing agents, with features like telephony support and global infrastructure for production use.
Is LiveKit available globally and what regions exist?
LiveKit operates globally with multiple regions (examples include locations in Australia, Brazil, France, Germany, India, Israel, Japan, Saudi Arabia, Singapore, South Africa, Switzerland, UAE, and US regions such as US Central, US East 1, US West). The platform emphasizes global infrastructure to support deployments worldwide.
Is there a free self-hosted option?
Yes. LiveKit offers an open-source framework that can be self-hosted for free, in addition to the cloud platform options.
.webp){kind=logo}
