改进生成对抗网络声码器，提升长时音频保真度与连贯性。
BemaGANv2: Discriminator Combination Strategies for GAN-based Vocoders in Long-Term Audio Generation
- 用AMP模块替代残差块，结合Snake激活函数更好建模周期性特征。
- 融合多包络判别器与多分辨率判别器，显著提升长时依赖建模能力。
- 系统评估多种判别器组合，在客观与主观评测中均表现优异。
本文提出BemaGANv2，一种面向高保真长时音频生成的先进生成对抗网络声码器，重点系统评估判别器组合策略。长时音频生成对文本到音乐（TTM）和文本到音频（TTA）系统至关重要，但维持长时间的时序连贯性、语调一致性和谐波结构仍是重大挑战。基于原始BemaGAN架构，BemaGANv2通过将生成器中的传统ResBlocks替换为抗混叠多周期性组合（AMP）模块，并在内部引入Snake激活函数，更有效地建模周期性结构。在判别器框架中，我们提出新型多包络判别器（MED），用于提取关键的时序包络特征以实现周期性检测；结合多分辨率判别器（MRD），该组合可更准确地建模音频中的长程依赖关系。我们系统评估了多种判别器配置，包括MSD+MED、MSD+MRD以及MPD+MED+MRD，采用客观指标（弗雷谢音频距离FAD、结构相似性指数SSIM、皮尔逊相关系数PCC、梅尔倒谱失真MCD、多分辨率STFT M-STFT、周期性误差Periodicity）和主观评估（MOS、SMOS）。为支持可复现性，论文提供详细的架构描述、训练配置及完整实现细节。代码、预训练模型与音频演示样本已公开：https://github.com/dinhoitt/BemaGANv2。
原文摘要 · Abstract (English)
This paper presents BemaGANv2, an advanced GAN-based vocoder designed for high-fidelity and long-term audio generation, with a focus on systematic evaluation of discriminator combination strategies. Long-term audio generation is critical for applications in Text-to-Music (TTM) and Text-to-Audio (TTA) systems, where maintaining temporal co- herence, prosodic consistency, and harmonic structure over extended durations remains a significant challenge. Built upon the original BemaGAN architecture, BemaGANv2 incorporates major architectural innovations by replacing traditional ResBlocks in the generator with the Anti-aliased Multi-Periodicity composition (AMP) module, which internally applies the Snake activation function to better model periodic structures. In the discriminator framework, we integrate the Multi-Envelope Discriminator (MED), a novel architecture we proposed, to extract rich temporal en- velope features crucial for periodicity detection. Coupled with the Multi-Resolution Discriminator (MRD), this com- bination enables more accurate modeling of long-range dependencies in audio. We systematically evaluate various discriminator configurations, including Multi-Scale Discriminator (MSD) + MED, MSD + MRD, and Multi-Period Discriminator (MPD) + MED + MRD, using objective metrics (Fréchet Audio Distance (FAD), Structural Similar- ity Index (SSIM), Pearson Correlation Coefficient (PCC), Mel-Cepstral Distortion (MCD), Multi-Resolution STFT (M-STFT), Periodicity error (Periodicity)) and subjective evaluations (MOS, SMOS). To support reproducibility, we provide detailed architectural descriptions, training configurations, and complete implementation details. The code, pre-trained models, and audio demo samples are available at: https://github.com/dinhoitt/BemaGANv2.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。