Audio classification methods work well when fine-tuned to reduced domains, such as musical instrument classification or simplified sound effects taxonomies. Classification methods cannot currently offer the detail needed in general sound recognition. A real-world-sound recognition tool would require thousands of classifiers, each specialized in distinguishing little details and a taxonomy that represents the real world. We describe the use of WordNet, a semantic network that organizes real world knowledge as the taxonomy backbone. In order to overcome the huge number of classifiers to distinguish an ever growing number of sounds, the recognition engine uses nearest-neighbor classifier with a database of isolated sounds unambiguously linked to WordNet concepts.
Click to purchase paper as a non-member or login as an AES member. If your company or school subscribes to the E-Library then switch to the institutional version. If you are not an AES member and would like to subscribe to the E-Library then Join the AES!
This paper costs $33 for non-members and is free for AES members and E-Library subscribers.