L3AC: Towards a Lightweight and Lossless Audio Codec

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Zhai, Linwei, Ding, Han, Zhao, Cui, wang, fei, Wang, Ge, Zhi, Wang, Xi, Wei
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908490440638464
author Zhai, Linwei
Ding, Han
Zhao, Cui
wang, fei
Wang, Ge
Zhi, Wang
Xi, Wei
author_facet Zhai, Linwei
Ding, Han
Zhao, Cui
wang, fei
Wang, Ge
Zhi, Wang
Xi, Wei
contents Neural audio codecs have recently gained traction for their ability to compress high-fidelity audio and provide discrete tokens for generative modeling. However, leading approaches often rely on resource-intensive models and complex multi-quantizer architectures, limiting their practicality in real-world applications. In this work, we introduce L3AC, a lightweight neural audio codec that addresses these challenges by leveraging a single quantizer and a highly efficient architecture. To enhance reconstruction fidelity while minimizing model complexity, L3AC explores streamlined convolutional networks and local Transformer modules, alongside TConv--a novel structure designed to capture acoustic variations across multiple temporal scales. Despite its compact design, extensive experiments across diverse datasets demonstrate that L3AC matches or exceeds the reconstruction quality of leading codecs while reducing computational overhead by an order of magnitude. The single-quantizer design further enhances its adaptability for downstream tasks. The source code is publicly available at https://github.com/zhai-lw/L3AC.
format Preprint
id arxiv_https___arxiv_org_abs_2504_04949
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle L3AC: Towards a Lightweight and Lossless Audio Codec
Zhai, Linwei
Ding, Han
Zhao, Cui
wang, fei
Wang, Ge
Zhi, Wang
Xi, Wei
Sound
Artificial Intelligence
68T07
I.2.m
Neural audio codecs have recently gained traction for their ability to compress high-fidelity audio and provide discrete tokens for generative modeling. However, leading approaches often rely on resource-intensive models and complex multi-quantizer architectures, limiting their practicality in real-world applications. In this work, we introduce L3AC, a lightweight neural audio codec that addresses these challenges by leveraging a single quantizer and a highly efficient architecture. To enhance reconstruction fidelity while minimizing model complexity, L3AC explores streamlined convolutional networks and local Transformer modules, alongside TConv--a novel structure designed to capture acoustic variations across multiple temporal scales. Despite its compact design, extensive experiments across diverse datasets demonstrate that L3AC matches or exceeds the reconstruction quality of leading codecs while reducing computational overhead by an order of magnitude. The single-quantizer design further enhances its adaptability for downstream tasks. The source code is publicly available at https://github.com/zhai-lw/L3AC.
title L3AC: Towards a Lightweight and Lossless Audio Codec
topic Sound
Artificial Intelligence
68T07
I.2.m
url https://arxiv.org/abs/2504.04949