Computer Encoding and Analysis of a Song Database for the Humpback Whale (Megaptera novaeangliae)
IAAAM Archive
Larry W. Junek

Humpback whale songs may be stored and analyzed on a computer using the system described in this paper. The computer system consists of a Macintosh computer with an audio signal digitizer, and software developed using HyperCard. Humpback whale songs may be divided temporally into themes, phrases, subphrases, and ultimately into discrete vocalization units. The entire song, which may last for hours, consist of a finite number of units repeated at different times at the same or a different pitch. The computer may take advantage of this repetition by encoding the entire song using relatively few individual units. Each unit is digitized into a computer storage database and given a unique ID number. The song itself is represented in the computer as a sequence of unit ID numbers, along with other information needed to reproduce the song, such as pitch and duration. In this system, units must be manually selected by a human operator using aural dictation techniques. Hence, a user friendly interface was developed to aid the dictation process. The song may then be analyzed in a number of ways. For example, a particular unit or sequence of units may be identified, or separate songs (say, from successive years) may be compared to identify deviations. If an entire song were digitized, it would take a very large amount of media storage, and analysis would be slow and difficult.

Speaker Information
(click the speaker's name to view other papers and abstracts submitted by this speaker)

Larry W. Junek


MAIN : Poster Session : Whale Song Database
Powered By VIN
SAID=27