FreeV：通过伪逆 Mel 滤波器为声码器带来免费午餐
FreeV: Free Lunch For Vocoders Through Pseudo Inversed Mel Filter
AI总结：
针对频域 GAN 声码器参数量大的问题，本文提出 FreeV，利用伪逆 Mel 滤波器初始化幅度谱并精简预测分支，以近半参数实现 1.8 倍提速且重合成质量更优。
摘要：
声码器从声学特征重建语音波形，并在现代 TTS 系统中发挥关键作用。近期，Vocos 和 APNet2 等频域 GAN 声码器发展迅速，在推理速度上超过时域模型，同时达到相当的音频质量。然而，这些频域声码器参数量较大，因此带来额外的内存负担。受 PriorGrad 和 SpecGrad 启发，我们采用伪逆来粗略估计幅度谱，并将其作为初始化。这种简单的初始化显著降低了声码器对参数的需求。基于 APNet2 以及我们精简后的幅度预测分支，我们提出了 FreeV。与其对应模型 APNet2 相比，FreeV 的推理速度提升 1.8 倍，参数量接近其一半。同时，FreeV 在重合成质量上优于 APNet2，标志着在追求实时、高保真语音合成方面迈出一步。代码和检查点可在 https://github.com/BakerBunker/FreeV 获取。
英文摘要：
Vocoders reconstruct speech waveforms from acoustic features and play a pivotal role in modern TTS systems. Frequent-domain GAN vocoders like Vocos and APNet2 have recently seen rapid advancements, outperforming time-domain models in inference speed while achieving comparable audio quality. However, these frequency-domain vocoders suffer from large parameter sizes, thus introducing extra memory burden. Inspired by PriorGrad and SpecGrad, we employ pseudo-inverse to estimate the amplitude spectrum as the initialization roughly. This simple initialization significantly mitigates the parameter demand for vocoder. Based on APNet2 and our streamlined Amplitude prediction branch, we propose our FreeV, compared with its counterpart APNet2, our FreeV achieves 1.8 times inference speed improvement with nearly half parameters. Meanwhile, our FreeV outperforms APNet2 in resynthesis quality, marking a step forward in pursuing real-time, high-fidelity speech synthesis. Code and checkpoints is available at: https://github.com/BakerBunker/FreeV
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
