Low-Resource Audio Codec (LRAC): 2025 Challenge Description

2025-10-27

AI Summary

Neural audio codecs face dual challenges in edge deployment: stringent resource constraints (computational efficiency, latency, bitrate) and acoustic robustness (e.g., noise, reverberation). To address this, we introduce the first public audio codec challenge explicitly designed to jointly optimize for low-resource efficiency and robustness. We construct a standardized multi-degradation dataset, lightweight baseline models, and a comprehensive evaluation framework. Methodologically, we propose a neural-hybrid codec architecture integrated with a speech enhancement frontend, enabling high-fidelity speech reconstruction at ultra-low bitrates (<1.6 kbps) while adhering to edge-device constraints. Our contributions are threefold: (1) establishing the first benchmark tailored to realistic edge scenarios; (2) advancing the co-optimization paradigm of coding and speech enhancement; and (3) open-sourcing a unified platform that attracted international participation, yielding multiple efficient and robust solutions—thereby providing technical and data foundations for ultra-low-bitrate voice communication and downstream tasks.

Technology Category

• Machine Learning: Learning on the Edge & Model Compression
• Cognitive Modeling & Cognitive Systems: Neural Spike Coding
• Search and Optimization: Mixed Discrete/Continuous Search

Application Category

• Economics, Online Markets and Human Computation: Research challenges in human and human-AI computation
• Responsible Web: Human-perceived consequences of algorithmic deployment on the web
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation

Abstract

While recent neural audio codecs deliver superior speech quality at ultralow bitrates over traditional methods, their practical adoption is hindered by obstacles related to low-resource operation and robustness to acoustic distortions. Edge deployment scenarios demand codecs that operate under stringent compute constraints while maintaining low latency and bitrate. The presence of background noise and reverberation further necessitates designs that are resilient to such degradations. The performance of neural codecs under these constraints and their integration with speech enhancement remain largely unaddressed. To catalyze progress in this area, we introduce the 2025 Low-Resource Audio Codec Challenge, which targets the development of neural and hybrid codecs for resource-constrained applications. Participants are supported with a standardized training dataset, two baseline systems, and a comprehensive evaluation framework. The challenge is expected to yield valuable insights applicable to both codec design and related downstream audio tasks.

Problem

• Developing neural codecs for low-resource edge deployment
• Enhancing codec robustness against acoustic distortions like noise
• Integrating codec performance with speech enhancement techniques

Innovation

• Neural codecs for low-resource edge deployment
• Hybrid designs resilient to acoustic distortions
• Standardized training dataset with evaluation framework