HearingDepth

HearingDepth是一款基于AI图像识别的声音创作软件。用户可以采集空间环境的深度，以及环境图像的物体分割。通过与屏幕的交互，空间深度和分割后的图像数据被转化为声音，作为一种声音实验。

在过去的几千年中，人类从未停止过对空间的探索，以应用在绘画及建筑等领域。如今，AI已经可以基于算法和数据，得到人类难以企及的精确度和复杂度。移动设备和可穿戴设备作为人体和智力的延伸，AI的结合无疑拓宽了它未来的可能性。人的身体带领 着”移动机器”去探索世界，认识世界，使其得到进化。搭载着"AI大脑" 的手机也不再仅仅是一台预编程的机器，而是拥有自我意识的第N器官。 最终到底是机器帮助人类进化?还是人类正向机器进化?已经成为一个非常有趣的问题。 在这个作品中，我主要利用AI视觉，对空间深度和图像分割进行识别，并用选择声音元素作为交互。探索AI如何看待世界?以及基于AI深度产生的声音，在真实空间中反射和接收的过程，会带来怎样的实验效应?

Humans never stop exploring the depth of space for more than 2,000 years in subjects such as painting and architecture. Nowadays, AI can already achieve precision and complexity result that humans can hardly reach based on algorithms and input data.
As mobile devices and wearable devices are the extensions of the human body and intelligence, the combination with AI will undoubtedly broaden its possibilities. The human body leads the “mobile machine” to explore the world, understand the world, and finally let it evolve. A mobile phone equipped with an AI brain is no longer just a pre-programmed machine, but a second human organ with self- awareness. This forces me to think about, are machines helping humans evolve? Or are humans evolving to machines? This is a question worth exploring.
In this work, I use AI vision to recognize the spatial depth and image segmentation, then select sound as interactive elements to make a response. How does AI see the world? And what kind of experimental effects will be brought about by the reflection and reception of sound generated by AI vision’s depth.