results: 对于十种声音效果的测试，NoiseBandNet得分高于四种变体的DDSP滤波器synthesizer，在九个评价类别中得分更高，表明NoiseBandNet可以生成具有时间和频率分辨率的各种听起来不同的声音效果。Abstract
Controllable neural audio synthesis of sound effects is a challenging task due to the potential scarcity and spectro-temporal variance of the data. Differentiable digital signal processing (DDSP) synthesisers have been successfully employed to model and control musical and harmonic signals using relatively limited data and computational resources. Here we propose NoiseBandNet, an architecture capable of synthesising and controlling sound effects by filtering white noise through a filterbank, thus going further than previous systems that make assumptions about the harmonic nature of sounds. We evaluate our approach via a series of experiments, modelling footsteps, thunderstorm, pottery, knocking, and metal sound effects. Comparing NoiseBandNet audio reconstruction capabilities to four variants of the DDSP-filtered noise synthesiser, NoiseBandNet scores higher in nine out of ten evaluation categories, establishing a flexible DDSP method for generating time-varying, inharmonic sound effects of arbitrary length with both good time and frequency resolution. Finally, we introduce some potential creative uses of NoiseBandNet, by generating variations, performing loudness transfer, and by training it on user-defined control curves.
摘要
<>translate into Simplified Chinese� controllable neural audio synthesis of sound effects is a challenging task due to the potential scarcity and spectro-temporal variance of the data. Differentiable digital signal processing (DDSP) synthesisers have been successfully employed to model and control musical and harmonic signals using relatively limited data and computational resources. Here we propose NoiseBandNet, an architecture capable of synthesising and controlling sound effects by filtering white noise through a filterbank, thus going further than previous systems that make assumptions about the harmonic nature of sounds. We evaluate our approach via a series of experiments, modelling footsteps, thunderstorm, pottery, knocking, and metal sound effects. Comparing NoiseBandNet audio reconstruction capabilities to four variants of the DDSP-filtered noise synthesiser, NoiseBandNet scores higher in nine out of ten evaluation categories, establishing a flexible DDSP method for generating time-varying, inharmonic sound effects of arbitrary length with both good time and frequency resolution. Finally, we introduce some potential creative uses of NoiseBandNet, by generating variations, performing loudness transfer, and by training it on user-defined control curves.Translation:控制可能的神经音频合成声效是一个挑战性的任务，因为声效数据的可能性和spectro-temporal variance很大。 diferenciable digital signal processing（DDSP）Synthesisers have been successfully employed to model and control musical and harmonic signals using relatively limited data and computational resources. 我们提议NoiseBandNet，一种可以通过filterbank filtering white noise来实现和控制声效的架构。这超过了之前的系统，它们假设声效的和谐性。我们通过一系列实验，模拟了踏步、雨天、陶艺、打击和金属声效。 Comparing NoiseBandNet的声音重建能力与四种DDSP滤波器处理的噪声合成器，NoiseBandNet在十个评价类别中得分高于其他四个， Establishing a flexible DDSP method for generating time-varying, inharmonic sound effects of arbitrary length with both good time and frequency resolution。最后，我们介绍了一些可能的创造性使用NoiseBandNet，如生成变化、卷积传递和用户定义的控制曲线。