Abstract
Neural audio synthesis methods now allow specifying ideas in natural language. However, these methods produce results that cannot be easily tweaked, as they are based on large latent spaces and up to billions of uninterpretable parameters. We propose a text-to-audio generation method that leverages a virtual modular sound synthesizer with only 78 parameters. Synthesizers have long been used by skilled sound designers for media like music and film due to their flexibility and intuitive controls. Our method, CTAG, iteratively updates a synthesizer's parameters to produce high-quality audio renderings of text prompts that can be easily inspected and tweaked. Sounds produced this way are also more abstract, capturing essential conceptual features over fine-grained acoustic details, akin to how simple sketches can vividly convey visual concepts. Our results show how CTAG produces sounds that are distinctive, perceived as artistic, and yet similarly identifiable to recent neural audio synthesis models, positioning it as a valuable and complementary tool.
Abstract (translated)
神经音频合成方法现在允许在自然语言中指定创意。然而，这些方法产生的结果很难进行微调，因为它们基于大型潜在空间和多达数十亿的无法解释的参数。我们提出了一种基于虚拟模块声音合成器的方法，该合成器只有78个参数。合成器一直以来都被有经验的音频设计师用于媒体如音乐和电影，因为它们的可伸缩性和直观的控制。我们的方法CTAG，通过迭代更新合成器的参数，产生高品质的文本提示音频渲染，这些音频可以轻松地被审查和微调。这种方式产生的声音也更抽象，捕捉到基本概念特征，与粗粒度的音频细节相似，就像简单的草图可以生动地传达视觉概念一样。我们的结果表明，CTAG产生的声音具有独特性，被认为具有艺术性，与最近神经音频合成模型产生类似的可识别性，将其定位为一种有价值且互补的工具。
URL
https://arxiv.org/abs/2406.00294
