A Short Example of an Enterprise Use Case (5 minutes)
An employee of a large enterprise has questions about his paid time off, the schedule for his laptop upgrade, and how to contact a colleague at another company. Fortunately, his company has automated these functions with conversational agents that can interoperate using Open Voice Interoperability protocols and pass along the conversation to each other. This short video illustrates the use case with an overview and working demo.
Build Interoperable Conversational Assistants with Open Voice Specifications
Part I: Specifications and technical introduction https://www.youtube.com/watch?v=E62Syvb5wJs
Part II: Hands on assistant development with Open Voice open source software https://www.youtube.com/watch?v=0LOeSG0eSfE&t=1955s
Hallucination Mitigation using Agentic AI Natural Language-Based Frameworks
Diego Gosmar and Deborah Dahl
http://arxiv.org/abs/2501.13946
This paper describes our work on using a pipeline of collaborating intelligent agents to mitigate hallucinations in LLM output. Besides demonstrating an innovative approach to hallucination mitigation, the paper also illustrates a use case for standard agent protocols beyond support for interactive applications.
AI Multi-Agent Interoperability Extension for Managing Multiparty Conversations
Diego Gosmar, Deborah A. Dahl, Emmett Coin, David Attwater
https://arxiv.org/abs/2411.05828
This paper describes our work on adding support for simultaneous multiparty conversations. It will also be presented at the AAAI workshop on Advancing LLM-Based Multi-Agent Collaboration in Philadelphia on March 4.
AI Multi-Agent Interoperability Extension for Managing Multiparty Conversations
Presented by Diego Gosamar
Authors: Diego Gosmar, Deborah A. Dahl, Emmett Coin and David Attwater.
Workshop on Natural Language for Artificial Intelligence (NL4AI, 8th edition)
Co-located @ AIxIA conference
November 25th – 28th, 2024, Bolzano, Italy
Developing Interoperable Conversational Assistants with the Open Voice Interoperability Initiative Specifications and Sandbox
Grace Hopper Celebration Open Source Day Hackathon, October 4, 2024, virtual
presented by Leah Barnes, Emmett Coin, and Deborah Dahl
This session described the Open Voice messaging protocol and introduced the Interoperability Sandbox implementation of the protocol. Attendees learned about the sandbox github repository (at https://github.com/open-voice-interoperability/open-voice-sandbox), and had a chance to develop their own standards-based chatbots using the sandbox.
Let’s Talk about Open Voice Interoperability
LF AI & Data Technical Advisory Council Talk, September 17, virtual, presented by Deborah Dahl
The session delved into the technical framework underpinning the interoperability initiative, outlining three primary types of messages that facilitate inter-assistant communication:
-
- Assistant Manifest
- Conversation Envelope
- Dialogue Events
Generative Conversational AI Interoperability
Open Source Summit Europe, Vienna, presented by Diego Gosmar
https://youtu.be/ZkxnEjUFY-k?si=9Y5NRX19XJXZJiFZ
This talk explores Conversational AI that works like the Web. In this talk, we explain the pivotal role of OVON Specs in the Open Voice Network’s Interoperability Initiative. We highlight how OVON Specs are crucial in crafting a universal, open-source AI voice interface for seamless cross-platform assistant integration.
Toward teams of AIs: Conversational assistants that converse with each other
AI for Developers Meeting: LF Edge and LF AI & Data (Hybrid), September 11 – September 13, 2024, presented by Deborah Dahl
This talk described emerging standards being developed by the Open Voice Interoperability Project of the LFAI and Data Foundation that will enable conversational assistants to interact with each other and transfer conversations to address use cases like these, based on a standard set of inter-assistant messages.
Conversational AI Multi-Agent Interoperability, Universal Open APIs for Agentic Natural Language Multimodal Communications Interoperability
Diego Gosmar, Deborah A. Dahl and Emmett Coin, 2024
https://arxiv.org/abs/2407.19438
Generative Conversational AI Interoperability
August 8, New York, NY, presented by Diego Gosmar, Emmett Coin, Deborah Dahl and Olga Howard
This one-day workshop presented the future of conversational AI in New York City. It was designed for professionals eager to master the integration and implementation of conversational assistants using cutting-edge technologies like Large Language Models (LLMs), Retrieval-Augmented Generation (RAG), and the pioneering OVON interoperability specifications.