Forager
Feel the magic of Machine Learning in this new experiment that combines over 44,000 sounds from Freesound, a data set created by Kyle McDonald and a touch of Cabbibo weirdness.
Run your hands through a cloud of sound, learn the shape of the computers mind, select your favorite audio delicacies and make a drum set from this decadent aural smorgasbord.
The mechanics may be simple ( navigate through a 3D map of sounds , select individual sounds to keep, and hit those sounds again to play them ), but because of absolutely massive size of the data set, the possibilities are endless!
Want to create your own piano out of synth notes? Go ahead!
How about a marimba of pops and crackles? Be my guest!
And if you are just the exploring type, the data is beautiful enough to begin with! There is no need to make an instrument, or play it, because in from of you lies a tool for discovering and understanding a beautiful algorithm, and even more beautiful data: Sound
This project would not be possible without the amazing community of FreeSound.org
As well as Kyle McDonald, who scraped audio from freesound.org & made t-SNE embeddings of the data!
Created during Unity Artist in Residency 2017
游戏讨论
Steam User 1
+1 免费 没中文 没成就 必须VR
《音频采集者》
在这款全新的实验作品中，感受机器学习的魅力。它融合了来自 Freesound 的超过44,000个声音、由 Kyle McDonald 创建的数据集，以及一丝 Cabbibo 式的奇思妙想。
用手穿过一片声音的云彩，了解计算机思维的形态，挑选你最喜欢的音频珍品，并用这场丰盛的声音盛宴打造一套属于自己的架子鼓。
操作机制或许很简单（在声音的3D地图中导航，选择你想要保留的声音，再次击中这些声音即可播放），但由于数据集的规模极其庞大，可能性是无穷无尽的！
想用合成音色制作你自己的钢琴？没问题！
想用爆裂声和咔嚓声做成一台马林巴琴？请便！
如果你只是喜欢探索的类型，数据本身就已经足够美妙了！你完全不需要制作乐器或演奏它——因为你面前的工具，正是用来发现和理解一个美丽算法，以及比算法更美的数据：声音。
这个项目的实现离不开 FreeSound org 这个了不起的社区。
同样也感谢 Kyle McDonald，他从 freesound org 抓取了音频数据，并生成了这些数据的 t-SNE 嵌入。
创作于 Unity 2017 驻场艺术家项目期间