Multi-Rate Bandwidth Extension by Token Completion in Neural Audio Codecs

2026-09-29

AI Summary

This study addresses the challenge of high-frequency reconstruction in audio bandwidth extension by reformulating the task as an audio token prediction problem. Methodologically, it introduces a novel neural audio codec guided by harmonic-percussive decomposition to facilitate disentangled representations, aligning its architecture with generative modeling and downstream tasks. A Transformer-based language model is then employed to process the discrete representations produced by this codec for high-frequency signal reconstruction. Experimental results demonstrate that the proposed approach achieves high-quality audio reconstruction in both objective metrics and subjective evaluations. Overall, this work establishes a new paradigm for bandwidth extension that effectively balances generative fidelity with downstream compatibility.

Abstract

Bandwidth extension, the task of reconstructing the high-frequency components of an audio signal from its low-passed counterpart, is a long-standing problem in audio processing. In this work, we extend recent advances in neural architectures by framing bandwidth extension as an audio token prediction problem. Specifically, we train a transformer-based language model on the discrete representations produced by a disentangled neural audio codec, where the disentanglement is guided by a Harmonic-Percussive decomposition of the input signals, highlighting spectral structures particularly relevant for bandwidth extension. Our approach introduces a novel codec design that explicitly accounts for the downstream token prediction task, enabling a more effective coupling between codec structure and transformer modeling. This joint design yields high-quality reconstructions of the original signal, as measured by both objective metrics and subjective evaluations. These results highlight the importance of aligning codec disentanglement and representation learning with the generative modeling stage, and demonstrate the potential of global, representation-aware design for advancing bandwidth extension.

Problem

• Bandwidth Extension
• Audio Processing
• Neural Audio Codecs
• Token Prediction

Innovation

• Bandwidth Extension
• Neural Audio Codec
• Token Prediction
• Harmonic-Percussive Decomposition
• Transformer