In this paper, we propose a novel conditional convolution network, named
location-variable convolution, to model the dependencies of the waveform
sequence. Different from the use of unified convolution kernels in WaveNet to
capture the dependencies of arbitrary waveform, the location-variable
convolution uses convolution kernels with different coefficients to perform
convolution operations on different waveform intervals, where the coefficients
of kernels is predicted according to conditioning acoustic features, such as
Mel-spectrograms. Based on location-variable convolutions, we design LVCNet for
waveform generation, and apply it in Parallel WaveGAN to design more efficient
vocoder. Experiments on the LJSpeech dataset show that our proposed model
achieves a four-fold increase in synthesis speed compared to the original
Parallel WaveGAN without any degradation in sound quality, which verifies the
effectiveness of location-variable convolutions.

本文提出了一种名为位置可变卷积的新型条件卷积网络来建模波形序列的依赖关系，其使用具有不同系数的卷积核对不同的波形区间进行卷积操作，该系数根据条件声学特征（如 Mel - 频谱图）预测，基于该方法，我们设计了 LVCNet 进行波形生成，并将其应用于 Parallel WaveGAN 中来设计更高效的声码器，LJSpeech 数据集上的实验结果表明，与原始的 Parallel WaveGAN 相比，我们提出的模型在不降低声音质量的前提下实现了四倍的合成速度提高，从而验证了位置可变卷积的有效性。

LVCNet: 高效的条件相关建模网络用于波形生成

LVCNet: Efficient Condition-Dependent Modeling Network for Waveform  Generation

We propose Parallel WaveGAN, a distillation-free, fast, and small-footprint
waveform generation method using a generative adversarial network. In the
proposed method, a non-autoregressive WaveNet is trained by jointly optimizing
multi-resolution spectrogram and adversarial loss functions, which can
effectively capture the time-frequency distribution of the realistic speech
waveform. As our method does not require density distillation used in the
conventional teacher-student framework, the entire model can be easily trained.
Furthermore, our model is able to generate high-fidelity speech even with its
compact architecture. In particular, the proposed Parallel WaveGAN has only
1.44 M parameters and can generate 24 kHz speech waveform 28.68 times faster
than real-time on a single GPU environment. Perceptual listening test results
verify that our proposed method achieves 4.16 mean opinion score within a
Transformer-based text-to-speech framework, which is comparative to the best
distillation-based Parallel WaveNet system.

本文提出了一种新型的声波生成方法，名为 Parallel WaveGAN，采用生成对抗网络作为模型，通过优化多分辨率的频谱图和对抗损失函数来训练非自回归 WaveNet，可以很好地捕捉现实语音信号的时频分布，在不需要常规师生框架中的密度蒸馏的情况下，整个模型容易训练且生成的语音高保真，适用于 Transformer-based 文本转语音框架 且与最佳的基于蒸馏的 Parallel WaveNet 系统相媲美。