Search Gestures Database

search call types db Search
A | B | C | D | E | F | G | H | I | J | K | L | M | N | O | P | Q | R | S | T | U | V | W | X | Y | Z

Design of new ElephantVoices audio database completed

A major milestone for ElephantVoices is the completion of a tailor-made database for the project’s thousands of recordings of elephant vocalizations. The database has been developed in close collaboration with a Kenyan IT student from the Jomo Kenyatta University of Agriculture and Technology, Phillip Nyamwaya, whom we have engaged to work with us on the project. Phillip has done a fantastic job, finding solutions to all of Joyce’s complicated requests.

The main user interface on EV Audio DatabaseThe database works a treat! The database will be a vital tool in the ongoing and time-consuming process of analyzing elephant calls, which will continue through 2005. While the actual measurement of audio files takes place in programs like Signal and Raven, ELEPHANTVOICES AUDIO DATABASE will be the very productive storage bank for the huge amount of data collected. The database, which links field notes (e.g. location, group size, call type, caller, behavioral context), measurements, audio files, video files, image files and spectrograms, will form the basis for understanding the acoustic repertoire of African savanna elephants. As work progresses we will make some of these data available on http://www.elephantvoices.org. Eventually we plan to make the best quality data in the database available on the web. Our hope is that we will be able to combine our work on African savanna elephants with the work of the Elephant Listening Project on African forest elephants and a future study of Asian elephants.

The database is built in MS Access with a Visual Basic interface. Scripts have been developed for different purposes related to updates, export of sound files and analysis. For example, by linking the ELEPHANTVOICES DATABASE to AERP’s 32-year demography database, the age of an individual elephant at the time of calling is automatically entered for each record. Thus, for instance, the database now includes calls by Emma at age 1 and calls by her as a mother at age 16. And our research assistant can now easily do part of the analysis in Amboseli (her favorite work place…), and by mailing a temporary storage file back to Nairobi, Joyce can easily add these measurements to their already existing records in the database.

Cheers, Petter/ElephantVoices