|
|
|
|
| Products Order On-Line Downloads News About Us Contact Links | ||
® Technology in the WildDespite all the impressive advances in computer technology, the most remarkable instrument for analyzing and identifying sound is still the human brain. The process is so automatic, effortless and instantaneous it goes unnoticed. We're not aware of what's happening as our brains process vast amounts of raw data, filter out background noise, focus on the sound of interest, and compensate for distance and variations in vocalizations.
No machine, not even the sophisticated, highly accurate Song Sleuth can match the processing power of the brain, especially in the wild. What's more, developing a digital device that has to work in the field is vastly more challenging than designing one to work in a sound studio or living room. Song Sleuth provides the utmost accuracy a machine is capable of by overcoming many of the unique challenges of operating in the wild.
First, while the field may be a peaceful place, it's also a noisy place. Winds blow, leaves rustle, branches snap, water laps, gushes or gurgles. Add to that a myriad of animal and bird sounds, distant traffic and airplane noise, even the rustle of your own clothing, and you're dealing with a multi-layered cacophony of sounds.
Song Sleuth overcomes much of the background noise problem in two ways. First, the innovative digital directional microphone array greatly reduces background noise by focusing on sounds directly in front of the array and attenuating, or diminishing, sounds from other directions. Thus, Song Sleuth is most accurate when aimed directly at, and as close as possible to, the source and away from other vocalizing birds.
Song Sleuth further reduces the effects of background noise by concentrating on the higher frequencies at which birds sing rather than the lower frequencies at which most noise occurs.
The second major challenge is the incredible range of individual and regional bird vocalizations. Not only do birds have an extensive repertoire of songs and calls, they have their own patterns and regional accents, like humans. Some sing softly, some loudly. Others don't sing very well, still others "mumble". Birds also produce calls, some as simple as one-note "chips" which even skilled birders find hard to distinguish. Adding to the confusion are the mimics like the Northern Mockingbird, with its uncanny ability to impersonate others.
Song Sleuth provides the most practical solution to the overwhelming variation in bird vocalizations by concentrating on the most common species and developing algorithms derived from multiple sources. Each software module includes only the birds you're most likely to hear in a specific geographic region or habitat, which greatly increases your chances for a "hit".
Each software module contains data for identifying the vocalizations of some sixty birds and is based on more than 500 individual authentic recordings representing 20 continuous hours of birdsongs. What's more, the algorithms developed using this data are capable of "learning" many individual and regional variations, which further increases the accuracy of Song Sleuth.
So while Song Sleuth is a highly accurate instrument, keep in mind it is not perfect. It identifies the most common birds, not every bird you may encounter in the field. And if a suspect bird sings a totally different tune or is a mimic, Song Sleuth won't identify it. Also keep in mind that the processing power of a human brain (and for that matter even a birdbrain) is superior to any computer. Until artificial intelligence and speech recognition technology matches brainpower, Song Sleuth is the invention that sets the standard for birdsong recognition.
Copyright © 2005-2008 Wildlife Acoustics, Inc. All Rights Reserved. Wildlife Acoustics and Song Sleuth are registered in the U.S. Patent and Trademark Office. Song Scope, Song Meter and Digital Birdsong Detective are trademarks of Wildlife Acoustics, Inc. Microsoft and Windows are either registered trademarks or trademarks of Microsoft Corporation in the United States and/or other countries. Apple and Mac OS are registered trademarks of Apple Computer, Inc. All other trademarks are the property of their respective owners. Patents pending.