SYNTHESIS AI UPDATE | AEC-Specific Video Transcription Model Milestone

We've been building an AEC-specific AI model for video transcriptions for the last several months. We ran the first benchmarking of the first generation of this model yesterday against a baseline transcription model from Microsoft Azure.

One of the areas the KA-built AEC-specific model really shines so far is acronyms and product names, and I've shared some of the highlights in the attached images.

We've partnered with our clients to build this model (and we have another on the way) through our Community AI program, an opt-in program that will empower Knowledge Architecture to create AEC-specific AI Infrastructure by training AI models on client content for the mutual benefit of participating firms, while also protecting confidentiality and intellectual property.

Once released into a public beta in June as part of our Synthesis Video 2 update, the AEC-specific video transcription model should help our clients make videos both more accessible and searchable on their intranets, as well as save our intranet champions hours and hours of editing video captions.

We're really excited to share this milestone progress update with all of you and will keep you posted as we approach public beta.

To learn more about our product roadmap, including Community AI, Synthesis Video 2, Next Generation Search, and Synthesis LMS, please visit our roadmap page.

SYNTHESIS AI ROADMAP | Next Generation Search, Video Captions, and Community AI

Knowledge Architecture has been working on two major AI projects for the last six months — Next Generation Search and Video Captioning.

At the same time, we have been developing an opt-in program called Community AI to create AEC-specific AI Infrastructure by training AI models on client content for the mutual benefit of participating firms.

We’re super excited about what we’re building and we hope you are too.

AI-Powered Enterprise Search for AEC Firms

Next Generation Search will bring AI-powered Enterprise Search to Synthesis.

In this video, we share our approach to both AI and Enterprise Search, including:

  • AI-Generated Search Summaries

  • Three Challenges with End-User Facing Large Language Models (LLMs)

  • Benefits of Retrieval-Augmented Generation (RAG)

  • Vector Search Basics

  • Benefits of Combining Keyword Search with Vector Search

  • Community AI Program Overview

  • Why We’re Building an AEC-Specific Embeddings Model

  • Enterprise Search Roadmap

  • Next Steps + Timeline for Public Beta

AI-Generated + AEC-Specific Video Captions

We’ll be releasing an update this spring that will add AI-generated captions, transcripts, and chapters, along with several other enhancements to Synthesis Native Video.

In this video, we go behind the scenes into:

  • Our AI-Powered Video Captioning Infrastructure

  • Community AI Program Overview

  • Benefits of our AEC-Specific Video Transcription Model

  • Next Steps + Release Timing

Introducing Our Community AI Program

Community AI is an opt-in program that will empower Knowledge Architecture to create AEC-specific AI Infrastructure by training AI models on client content for the mutual benefit of participating firms, while also protecting confidentiality and intellectual property.

In the pursuit of researching two use cases for AI — Next Generation Search and Video Captions — we discovered a unique opportunity to work with our AEC client community to achieve more relevant search results and better captioning.

Thus, Community AI was born.