Overview of Special Issue on Neural Speech and Audio Coding (video)

This webinar, part of the IEEE JSTSP webinar series, will focus on a recent special issue on Neural Speech and Audio Coding (NSAC}, a rapidly emerging research field at the intersection of signal processing and artificial intelligence (AI). Recent advances in deep neural networks and data-driven methodologies have opened new possibilities for unprecedented coding gain, which is far beyond the capabilities of traditional codecs. The special issue focuses on the challenges, insights, and opportunities that arise when applying AI-based methods to speech and audio coding, emphasizing novel architectures, inference-time efficiency, low-delay processing, and robustness to adverse acoustic conditions. The webinar will start with a brief introduction to the area of NSAC, followed by six presentations of technical papers appearing in the special issue.