Conversational AI Multi-Agent Interoperability, Universal Open APIs for Agentic Natural Language Multimodal Communications

Diego Gosmar,Deborah A. Dahl,Emmett Coin
2024-07-28
Abstract:This paper analyses Conversational AI multi-agent interoperability frameworks and describes the novel architecture proposed by the Open Voice Interoperability initiative (Linux Foundation AI and DATA), also known briefly as OVON (Open Voice Network). The new approach is illustrated, along with the main components, delineating the key benefits and use cases for deploying standard multi-modal AI agency (or agentic AI) communications. Beginning with Universal APIs based on Natural Language, the framework establishes and enables interoperable interactions among diverse Conversational AI agents, including chatbots, voicebots, videobots, and human agents. Furthermore, a new Discovery specification framework is introduced, designed to efficiently look up agents providing specific services and to obtain accurate information about these services through a standard Manifest publication, accessible via an extended set of Natural Language-based APIs. The main purpose of this contribution is to significantly enhance the capabilities and scalability of AI interactions across various platforms. The novel architecture for interoperable Conversational AI assistants is designed to generalize, being replicable and accessible via open repositories.
Artificial Intelligence,Human-Computer Interaction
What problem does this paper attempt to address?
The main problem that this paper attempts to solve is the lack of interoperability among current Conversational AI agents. With the rapid growth in the number of Chatbots and Voicebots, effective collaboration among these agents is becoming increasingly important. However, due to technological fragmentation and the lack of unified standards, communication and collaboration between different agents face complexity and inefficiency problems. Therefore, the paper proposes a new architecture based on open standards, aiming to achieve interoperability among different Conversational AI agents by using a common API (Application Programming Interface), thereby enhancing the cross - platform interaction capabilities and scalability. Specifically, the paper introduces a new architecture proposed by the Open Voice Interoperability Initiative. This architecture allows different types of Conversational AI agents (such as Chatbots, Voicebots, Video bots, and human agents) to interoperate through a natural - language - based common API. In addition, the paper also introduces a new discovery specification framework for efficiently finding agents that provide specific services and obtaining accurate information about these services through standard manifest publication. This new method not only simplifies the interaction between agents but also supports multi - modal AI agent communication, enhancing the flexibility and efficiency of agent interaction. In summary, the main objective of the paper is to solve the existing interoperability problems among current Conversational AI agents by proposing a new, open - standard - based interoperability framework, in order to promote the development of a more efficient and flexible AI interaction ecosystem.