Pritham and Sameer from our London office recently attended an insightful event focused on the integration of artificial intelligence in acoustics. The conference was organized by the Institute of Acoustics (IOA) and the UK Acoustics Network (UKAN), and featured a variety of compelling talks, including a case study on using AI for audio recognition in construction noise, and the application of text-to-image generators for initial project conceptualization.
Additionally, advancements in room acoustic simulation software and the training of AI models with edge case auralizations were explored. The event also provided updates from initial attempts at acoustic performance prediction using learning models and the potential benefit in moving from “blackbox” to “greybox” models that incorporate physics parameters alongside the original blackbox learning models. The discussions highlighted the exciting potential of AI in our field, while also underscoring the need for thoughtful use due to the associated energy and carbon costs.
