IntroductionOur lab would like to understand the grammar and meaning of bird songs. Bird songs are not random; they do have rules for putting notes and phrases together. We would like to characterize those rules. Recent advances in sensor arrays, computation, and computational linguistics finally make this long-sought goal achievable. The approach taken by our laboratory is to: (1) collect a large amount of bird song recordings from hand-held or acoustic sensor arrays in a variety of natural settings; (2) process the data by software, some of which is recent and some of which we are developing with new advances in localizing sources with beamforming, then filtering out noise, identifying events of interest, and then classifying them according to species and individual, combining that with behavioral observations when possible; (3) this information/knowledge is then stored in a large database that is shared among collaborating research groups; and (4) analyzing the recordings by computational-linguistic tools to identify the syntax of the songs. Combined that linguistic understanding with information about the context in which it occurred, our aim to explore new software methods to identify the meaning of those songs. We collaborate with ecologists, engineers, linguists and artists.
Our work is contributing to a profound transformation that is already underway: the recognition of very sophisticated signaling strategies and syntactic structures in non-human species. The new tools and methods for collecting and analyzing bird song now allow a level of observation that previously would not have been possible. We are now collecting truly vast amounts of data from previously inaccessible settings and subjecting data to previously undiscovered sophisticated structural analyses. Our work may also be transformational to computational linguistics if the natural world beyond humans were shown to have languages that are radically different from our own (as seems quite likely).
Some selected references about this work.
2012 Sasahara K, Cody ML, Cohen D, Taylor CE. Structural Design Principles of Complex Bird Songs: A Network-Based Approach. PLoS ONE 7(9): e44436. doi:10.1371/journal.pone.0044436
2014 Zhang, J, G. Kossan, R. Hedley, R. Hudson, C. E. Taylor, K. Yao and M. Bao. Fast 3D AML-based Bird Song Estimation. Unmanned Systems 2:249. DOI: 10.1142/S2301385014400044.
2014 Arriaga, J. G., Sanchez, H., Hedley, R., Vallejo, E. E., & Taylor, C. E. (2014). Using Song to Identify Cassinâs Vireo Individuals. A Comparative Study of Pattern Recognition Algorithms. In Pattern Recognit ion (pp. 291-300). Springer International Publishing.
2013 Tan, Lee Ngee, George Kossan, Martin L. Cody, Charles E. Taylor, Abeer Alwan. A sparse representation-based classifier for in-set bird phrase verification and classification with limited training data. The 38th International Conference on Acoustics, Speech, and Signal Processing (ICASSP), Vancouver, Canada, May 16-31, 2013.
2011 Kirschel, A. N. G., M. L. Cody, Z. Harlow, V. J. Promponas, E. E. Vallejo and C. E. Taylor. Territorial dynamics of Mexican Antthrushes revealed by individual recognition of their songs. Ibis. 153:255-268.
pdf versions of these papers, along a complete listing of our papers in this area may be found at /taylor0.biology.ucla.edu/al.publist.htm 605-434-6228.
In the past our lab was also researching computing tools for genetically modified mosquitoes to control and Malaria and Anopheles gambiae. See links in the left column of this web site for information and our publications on malaria and Anopheles.
June 2016Victoria Vesna, Takashi Ikegami and their students have made some impressive outreach installations in New York and Japan during the past year. A brief overview and nice videos are at (760) 929-2597
October 2014Victoria Vesna and the Art | Design group have fixed up the project web page, with some nice interviews and scenes from the working area near Volcano HTTP.
Complementing the video for the Art | Sci Nanotechnology day with high scholl students, HTTPS, thereis now a video for the Birdsong Diamond installation at HTTPS.
There will be a new version of Birdsong Diamond, indoors this time, on November 6, coupled with a LASER evening, see HTTPS.
September 2014George Kossan will be moving from the lab to a position at the Bank of America. His duties will be covered by Carolee Caffrey, who is joining us as a researcher, from Hofstra and Harvard.
Dent Earl has accepted a position at Google, and will be joining their Maps effort in October.
Hector Sanchez has made an oline version of his PajaroLoco program to analyze sound sequence recordings at HTTPS.
August 2014Julio Arriaga has made an oline version of his database of our sound sequence recordings at HTTPS.
We are still writing up a description. This is meant to be a community resource for those working on the syntax of animal vocalizations.
If you have comments or suggestions please contact him or CT.
July 2014Takashi Ikegami and Mitzuki Ota, with their students Atsushi Masumori, Itsuki Doi, and Norihiro Maruyama, with help from Victoria Vesna presented a sound installation, "Birdsong Diamond", in the Murphy Sculpture Garden. The poster is at jpg and the talk at Alife in New York City is at PDF.
Some scenes from this are in the video for the Art | Sci Nanotechnology day, HTTPS. There should be a video just for the installation available before long.
Victoria Vesna's group has been working on a new web site for the bird project. It is still not fully functional, but a version is at 9372726396
June 2014Past student John Marshall has a new position as Assistant Professor at UC Bekeley