Multi-Agent Architecture Best Practices Whitepaper Series

Today, we are excited to announce the Voice Interoperability Initiative (VII) Architecture Best Practices whitepaper series, and release the first whitepaper entitled “Foundational Concepts”. This whitepaper provides recommendations and architectural suggestions for consideration when building products that support multiple simultaneous voice agents. It is a deeper dive into the architectural considerations to achieve the recommendations outlined in the Multi-Agent Design Guide.

VII is an initiative committed to providing customers with choice and flexibility through multiple, interoperable voice services. The initiative is built around a shared belief that voice services should work seamlessly alongside one another on a single device, and that voice-enabled products should be designed to support multiple simultaneous wake words. We now have more than 80 members from diverse industries and backgrounds- across consumer electronics brands, automotive manufacturers, telecommunications operators, hardware solutions providers, and systems integrators. The group is working to make history by utilizing our respective expertise and experiences in support of this common goal.

Amazon is excited to publish the first whitepaper to provide support for device makers to build multi-agent devices that delight customers. We believe the concepts advanced in this whitepaper will be useful for achieving interoperability patterns such as those considered in the Multi-Agent Design Guide. Whether you are a technical architect, an engineer for a device maker, or a voice agent developer, this whitepaper will help you:

• Understand important concepts for creating experiences that align with the Multi Agent Design Guide
• Understand the implications of creating frustration-free experiences
• Design scalable solutions for building multi-agent experiences that support Universal Device Commands and Agent Transfers

We are looking to release more whitepapers later this year which will continue to build upon the concepts laid out in the Foundational Concepts whitepaper to dive deep into specific topics for consideration in connection with devices with multiple simultaneously active agents.