Bigvgan: A Universal Neural Vocoder With Large-scale Training

Abstract

Despite recent progress in generative adversarial network (GAN)-based vocoders, where the model generates raw waveform conditioned on acoustic features, it is challenging to synthesize high-fidelity audio for numerous speakers across various recording environments. In this work, we present BigVGAN, a universal vocoder that generalizes well for various out-of-distribution scenarios without fine-tuning. We introduce periodic activation function and anti-aliased representation into the GAN generator, which brings the desired inductive bias for audio synthesis and significantly improves audio quality. In addition, we train our GAN vocoder at the largest scale up to 112M parameters, which is unprecedented in the literature. We identify and address the failure modes in large-scale GAN training for audio, while maintaining high-fidelity output without over-regularization. Our BigVGAN, trained only on clean speech (LibriTTS), achieves the state-of-the-art performance for various zero-shot (out

Stats

• citations0
• S2 citations—
• github stars1212
• HF likes0
• heat score6.17
• arxiv keylee2022bigvgan

Related papers

• Vocgan: A High-fidelity Real-time Vocoder With A Hierarchically-nested Adversarial Network (2020)12.54
• Bigvsan: Enhancing Gan-based Neural Vocoders With Slicing Adversarial Network (2023)12.43
• Snakegan: A Universal Vocoder Leveraging DDSP Prior Knowledge And Periodic Inductive Bias (2023)4.52
• Vnet: A Gan-based Multi-tier Discriminator Network For Speech Synthesis Vocoders (2024)2.26
• DSPGAN: A Gan-based Universal Vocoder For High-fidelity TTS By Time-frequency Domain Supervision From DSP (2022)9.03
• Universal Melgan: A Robust Neural Vocoder For High-fidelity Waveform Generation In Multiple Domains (2020)0.00
• EVA-GAN: Enhanced Various Audio Generation Via Scalable Generative Adversarial Networks (2024)0.00
• Singgan: Generative Adversarial Network For High-fidelity Singing Voice Generation (2021)10.61