Neural Spectral Band Generation for Audio Coding

2025-06-07

AI Summary

Bandwidth extension (BWE) aims to reconstruct missing high-frequency components in bandwidth-limited audio signals. Conventional spectral band replication (SBR) offers controllability but suffers from poor generalization, whereas existing blind deep neural network (DNN)-based approaches lack explicit prior guidance and thus exhibit limited reconstruction fidelity. This paper proposes a novel parametric non-blind BWE paradigm: for the first time, a DNN is embedded within a traditional audio coding framework—low-frequency–driven side information is extracted at the encoder, and high-frequency spectral bands are conditionally synthesized at the decoder. This design synergistically integrates SBR’s controllability with DNNs’ powerful representation learning capability. The method overcomes the performance ceiling of purely blind BWE, achieving substantial improvements in high-frequency reconstruction quality for both speech and music. Objective evaluations show superior PESQ and STOI scores compared to conventional SBR and state-of-the-art blind DNN methods.

Technology Category

• Machine Learning: Deep Generative Models & Autoencoders
• Cognitive Modeling & Cognitive Systems: Neural Spike Coding
• Natural Language Processing: Speech

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Search and Retrieval-Augmented AI: Multilingual and cross-lingual Web search
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

Audio bandwidth extension is the task of reconstructing missing high frequency components of bandwidth-limited audio signals, where bandwidth limitation is a common issue for audio signals due to several reasons, including channel capacity and data constraints. While conventional spectral band replication is a well-established parametric approach to audio bandwidth extension, the SBR usually entails coarse feature extraction and reconstruction techniques, which leads to limitations when processing various types of audio signals. In parallel, numerous deep neural network-based audio bandwidth extension methods have been proposed. These DNN-based methods are usually referred to as blind BWE, as these methods do not rely on prior information extracted from original signals, and only utilize given low frequency band signals to estimate missing high frequency components. In order to replace conventional SBR with DNNs, simply adopting existing DNN-based methodologies results in suboptimal performance due to the blindness of these methods. My proposed research suggests a new approach to parametric non-blind bandwidth extension, as DNN-based side information extraction and DNN-based bandwidth extension are performed only at the front and end of the audio coding pipeline.

Problem

• Reconstruct missing high frequency audio components
• Improve conventional spectral band replication limitations
• Propose DNN-based parametric non-blind bandwidth extension

Innovation

• DNN-based side information extraction
• Parametric non-blind bandwidth extension
• Integrated audio coding pipeline