arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2006.09266eess.AScs.SD
收藏

使用生成对抗网络比较音频合成的表示

Comparing Representations for Audio Synthesis Using Generative Adversarial Networks

Sony Computer Science Laboratories(索尼计算机科学实验室);Télecom Paris(巴黎电信学院);Institut Polytechnique de Paris(巴黎理工学院)

Javier Nistal, Stefan Lattner, Gaël Richard

摘要:

在本文中,我们比较了不同的音频信号表示,包括原始音频波形和各种时频表示,用于生成对抗网络(GANs)的音频合成任务。我们在NSynth数据集的一个子集上进行实验。该架构遵循基准Progressive Growing Wasserstein GAN。我们以完全无条件的方式进行实验,并在音高信息上调节网络。我们利用标准指标对生成的材料进行定量评估,以评估生成模型,并比较训练和采样时间。我们表明,复值以及幅度和瞬时频率的短时傅里叶变换达到最佳效果,并产生快速生成和反转时间。用于特征提取、训练和评估模型的代码可在线获得。

英文摘要:

In this paper, we compare different audio signal representations, including the raw audio waveform and a variety of time-frequency representations, for the task of audio synthesis with Generative Adversarial Networks (GANs). We conduct the experiments on a subset of the NSynth dataset. The architecture follows the benchmark Progressive Growing Wasserstein GAN. We perform experiments both in a fully non-conditional manner as well as conditioning the network on the pitch information. We quantitatively evaluate the generated material utilizing standard metrics for assessing generative models, and compare training and sampling times. We show that complex-valued as well as the magnitude and Instantaneous Frequency of the Short-Time Fourier Transform achieve the best results, and yield fast generation and inversion times. The code for feature extraction, training and evaluating the model is available online.

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。

↑