But does it work?
Sent: Friday, 18 July 2014 3:43 AM
Subject: Re: [Nature Recordists] progress in automated birdsong recognition=
That's way cool, Dan. Thanks for bringing this to our attention.
On Jul 17, 2014, at 6:33 PM, Dan Dugan <=
.com> [naturerecordists] <<naturerec=
Forwarded to me (Dan Dugan):
Scientists from Queen Mary University of London have found a successful way=
of identifying bird sounds from large audio collections, which could be us=
eful for expert and amateur bird-watchers alike.
The analysis used recordings of individual birds and of dawn choruses to id=
entify characteristics of bird sounds. It took advantage of large datasets =
of sound recordings provided by the British Library Sound Archive, and onli=
ne sources such as the Dutch archive called Xeno Canto.
Publishing in the journal PeerJ, the authors describe an approach that comb=
ines feature-learning -- an automatic analysis technique -- and a classific=
ation algorithm, to create a system that can distinguish between which bird=
s are present in a large dataset.
"Automatic classification of bird sounds is useful when trying to understan=
d how many and what type of birds you might have in one location," commente=
d lead author Dr Dan Stowell from QMUL's School of Electronic Engineering a=
nd Computer Science and Centre for Digital Music.
Dr Stowell was recently awarded a five-year fellowship from the Engineering=
and Physical Sciences Research Council (EPSRC) to develop computerised pro=
cesses to detect multiple bird sounds in large sets of audio recordings.
"Birdsong has a lot in common with human language, even though it evolved s=
eparately. For example, many songbirds go through similar stages of vocal l=
earning as we do, as they grow up, which makes them interesting to study. F=
rom them we can understand more about how human language evolved and social=
organisation in animal groups," said Dr Stowell.
He added: "The attraction of fully automatic analysis is that we can create=
a really large evidence base to address these big questions."
The classification system created by the authors performed well in a public=
contest using a set of thousands of recordings with over 500 bird species =
from Brazil. The system was regarded as the best-performing audio only clas=
sifier, and placed second overall out of entries from 10 research groups in=
The researchers hope to drill down into more detail for their next project.
Dr Stowell says, "I'm working on techniques that can transcribe all the bir=
d sounds in an audio scene: not just who is talking, but when, in response =
to whom, and what relationships are reflected in the sound, for example who=
is dominating the conversation."
The above story is based on materials<http://www.qmul.ac.uk/media/news/item=
s/se/136151.html> provided by Queen Mary, University of London<http://www.q=
mul.ac.uk/>. Note: Materials may be edited for content and length.
1. Stowell D, Plumbley MD. Automatic large-scale classification of bird s=
ounds is strongly improved by unsupervised feature learning. PeerJ, 2014 DO=
Glen Ellen, CA 95442
TED Global talk (12Jun13): http://www.ted.com/talks/bernie_krause_the_voice=