We are planning a project to provide human-generated audio annotations for the ~800,000 videos in the YFCC100M dataset. This will be an open-source resource for the multimedia community, as part of the Multimedia Commons, http://www.multimediacommons.org.

So we need your input about how to make it most useful to YOU! This survey will help us get an idea what the annotation needs and priorities of the research community are, so we can decide where to focus our efforts for full- and partial-corpus annotation in AudioNet.

Feel free to skip questions -- we know you’re busy, and we’d rather have some answers than none! Most important is the first "Big Questions" section.