SLPNarrative ReviewTrends in hearing2021

New Avenues in Audio Intelligence: Towards Holistic Real-life Audio Understanding.

Björn Schuller, Alice Baird, Alexander Gebhard, Shahin Amiriparian, Gil Keren, Maximilian Schmitt, Nicholas Cummins

PMID 34751066

WHAT IT FOUND

This paper proposes a theoretical framework for AI that can separate and interpret overlapping sounds in real-world environments, such as distinguishing speech from noise in hearing aids.

It describes potential future applications but reports no clinical evidence or tested outcomes.

What this paper is

This is a theoretical overview and proposal for a new type of artificial intelligence system designed to understand complex audio environments. It reviews existing computer science literature and outlines a conceptual framework for future software development. It does not report on any clinical trials, patient outcomes, or tested interventions, so there is nothing here to change your practice on.

Read it on PubMed →