That=92s way cool, Dan. Thanks for bringing this to our attention.
On Jul 17, 2014, at 6:33 PM, Dan Dugan [naturerecordists] =
> Forwarded to me (Dan Dugan):
> Scientists from Queen Mary University of London have found a successful w=
ay of identifying bird sounds from large audio collections, which could be =
useful for expert and amateur bird-watchers alike.
> The analysis used recordings of individual birds and of dawn choruses to =
identify characteristics of bird sounds. It took advantage of large dataset=
s of sound recordings provided by the British Library Sound Archive, and on=
line sources such as the Dutch archive called Xeno Canto.
> Publishing in the journal PeerJ, the authors describe an approach that co=
mbines feature-learning -- an automatic analysis technique -- and a classif=
ication algorithm, to create a system that can distinguish between which bi=
rds are present in a large dataset.
> "Automatic classification of bird sounds is useful when trying to underst=
and how many and what type of birds you might have in one location," commen=
ted lead author Dr Dan Stowell from QMUL's School of Electronic Engineering=
and Computer Science and Centre for Digital Music.
> Dr Stowell was recently awarded a five-year fellowship from the Engineeri=
ng and Physical Sciences Research Council (EPSRC) to develop computerised p=
rocesses to detect multiple bird sounds in large sets of audio recordings.
> "Birdsong has a lot in common with human language, even though it evolved=
separately. For example, many songbirds go through similar stages of vocal=
learning as we do, as they grow up, which makes them interesting to study.=
From them we can understand more about how human language evolved and soci=
al organisation in animal groups," said Dr Stowell.
> He added: "The attraction of fully automatic analysis is that we can crea=
te a really large evidence base to address these big questions."
> The classification system created by the authors performed well in a publ=
ic contest using a set of thousands of recordings with over 500 bird specie=
s from Brazil. The system was regarded as the best-performing audio only cl=
assifier, and placed second overall out of entries from 10 research groups =
in the competition.
> The researchers hope to drill down into more detail for their next projec=
> Dr Stowell says, "I'm working on techniques that can transcribe all the b=
ird sounds in an audio scene: not just who is talking, but when, in respons=
e to whom, and what relationships are reflected in the sound, for example w=
ho is dominating the conversation."
> Story Source:
> The above story is based on materials provided by Queen Mary, University =
of London. Note: Materials may be edited for content and length.
> Journal Reference:
> Stowell D, Plumbley MD. Automatic large-scale classification of bird soun=
ds is strongly improved by unsupervised feature learning. PeerJ, 2014 DOI:1=
Glen Ellen, CA 95442
TED Global talk (12Jun13): http://www.ted.com/talks/bernie_krause_the_voice=