Saved in:
| Main Authors: | , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2506.09487 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866910045505060864 |
|---|---|
| author | Park, Taesoo Jeong, Mungwi Park, Mingyu Kim, Narae Kim, Junyoung Kim, Mujung Yoo, Jisang Lee, Hoyun Kim, Sanghoon Kwon, Soonchul |
| author_facet | Park, Taesoo Jeong, Mungwi Park, Mingyu Kim, Narae Kim, Junyoung Kim, Mujung Yoo, Jisang Lee, Hoyun Kim, Sanghoon Kwon, Soonchul |
| contents | 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. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2506_09487 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | BemaGANv2: Discriminator Combination Strategies for GAN-based Vocoders in Long-Term Audio Generation Park, Taesoo Jeong, Mungwi Park, Mingyu Kim, Narae Kim, Junyoung Kim, Mujung Yoo, Jisang Lee, Hoyun Kim, Sanghoon Kwon, Soonchul Sound Artificial Intelligence Machine Learning Logic in Computer Science Audio and Speech Processing I.2.6; H.5.5; I.5.1 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. |
| title | BemaGANv2: Discriminator Combination Strategies for GAN-based Vocoders in Long-Term Audio Generation |
| topic | Sound Artificial Intelligence Machine Learning Logic in Computer Science Audio and Speech Processing I.2.6; H.5.5; I.5.1 |
| url | https://arxiv.org/abs/2506.09487 |