Generative Conversational AI Interoperability
summary
OVON Specs define a JSON Universal API JSON envelope and a simple manifest discovery format that let heterogeneous assistants exchange utterances, whispers and invites as a state machine (Idle→Ready→…); a Python sandbox (open on GitHub, listens on port 60002) implements the spec, runs multi-agent demos (Cassandra florist, restaurant, post office), and has been trialed by Estonia’s government. Claims: cross-platform agent handoffs, RAG integration, and scalable discovery — spec is lightweight JSON, extensible, and already runnable.
AI-written summary · about
chapters (9)
- 0:00Introduction
- 2:54AI models overview
- 6:54What is conversational AI
- 11:12Why interoperability needed
- 14:04Open Voice Interoperability
- 15:08Universal API for AI agents
- 18:21Live demo: AI interoperability
- 25:25Team, resources, contribute
- 27:28Q&A
related talks
31mGenerative Conversational AI Interoperability
42mConversational Assistants that Converse with Each Other
28mOpey, the AI for API Discovery and Future of Fintech
23mOFP: Agentic AI Standard for Hallucination Mitigation and Multi-Agent Collaboration
24mLightning Talk: Building the AIverse: Standardizing Autonomous AI Agents for a Decent...
23mBuilding with OpenAI: What's Ahead
43mA2A in 50 Minutes: Building Interoperable AI Agents
47mStructured Outputs in OpenAI: Building Reliable JSON-Based AI Integrations