Vocalsketch Data Set V1.0.4

This data set contains thousands of vocal imitations of a large set of diverse sounds. These imitations were collected from hundreds of contributors via Amazon's Mechanical Turk website. The data set also contains data on hundreds of people's ability to correctly label these vocal imitations, also collected via Amazon's Mechanical Turk. This data set will help the research community understand which audio concepts can be effectively communicated with this approach. We have released this data so the community can study the related issues and build systems that leverage vocal imitation as an interaction modality, such as search engines that can be queried by vocally imitating the desired sound. <strong>This data set is a supplement to a paper. Please cite the following paper to reference this data set in a publication:</strong> Cartwright, M., Pardo, B. VocalSketch: Vocally Imitating Audio Concepts. In <em>Proceedings of ACM Conference on Human Factors in Computing Systems</em> (2015). http://dx.doi.org/10.1145/2702123.2702387 See https://github.com/interactiveaudiolab/VocalSketchDataSet for the latest updates to this data set. Interactive Audio Lab: http://music.eecs.northwestern.edu