Real-Time Audio Source Separation on iOS Devices

Explore real-time audio source separation on iOS devices in this 30-minute conference talk from the Audio Developer Conference (ADC22). Dive into the challenges of implementing low-latency source separation for musicians, focusing on djing and music learning applications. Learn about state-of-the-art deep learning-based source separation algorithms, model optimization techniques for iOS mobile platforms using Core ML and ANE, and low-level integration for real-time use cases. Gain insights from audio developers Adrien Ferrigno and Clément Tabary as they share their expertise in tackling the complexities of bringing this technology to mobile devices, potentially revolutionizing how musicians interact with music in real-time settings.