Replaying recordings of animal vocalizations is always a thrill. The sounds of the wild can be beautiful, stirring, even shocking. But when it comes to the scientific study of these sound recordings – known as bioacoustics – listening to sound is, unfortunately, one of the least useful ways to perform a scientific analysis.
Analyzing vocalizations by ear can be time-consuming. And how would you know precisely when a call occurred, at what amplitude or frequency, and under what circumstances?
That’s why sound visualization tools — namely oscillograms and spectrograms —are so crucial. By creating a tangible view of wildlife sounds, you can see vocalizations as they occur through time.
Let’s explore these tools and how they can be used to get the precision you need as a wildlife biologist.
An oscillogram is a visual representation of sound data which explains the relationship between time and amplitude.
The vertical axis on an oscillogram represents amplitude while the horizontal axis shows seconds of time. Each point along the waveform represents relative amplitude in decibels – meaning, the tool allows you to see and measure how amplitude changes over time from loud to soft.
A spectrogram provides a view of amplitude and frequency over time.
Like an oscillogram, it shows elapsed time in the same left-to-right representation, but with the added dimension of frequency.
Frequency is displayed on the vertical axis in hertz, and amplitude is represented by variations in color that show the intensity of the vocalization that occurred.
While you might be quick to assume that the spectrogram is superior due to its ability to show the added dimension of frequency, this isn’t always the case.
The oscillogram allows you to get a closer look at points of interest. For example, want a precise measurement of decibel changes in amplitude? You will get the most granular view from the oscillogram.
If you were to try to zoom down to sample level and retrieve the same information from a spectrogram, it would be much more difficult to see the same level of detail.
But that’s not to say the spectrogram is inferior either. In fact, it’s often a favorite among wildlife biologists. This is where the magic of the wild seems to come to life on the screen to tell a story, illustrating a dimensional view of sound.
With the added benefit of seeing frequency on a spectrogram, you can more easily discern between vocalizations at high and low frequencies. Vocal frequencies help you more quickly identify your species of study, so you can narrow your focus as you parse through your recordings.
For instance, you know that songbird frequencies would be higher than owl frequencies, so if you are studying songbirds, the higher frequency vocalizations would be your focus.
In truth, neither the oscillogram nor spectrogram is necessarily better than the other – they are simply different tools for different jobs. You can find different information in each, which can be equally valuable depending on the requirements of your research.
When making an oscillogram versus spectrogram comparison, it’s not a question of what the visualization tools cannot do, but rather a question of what they can do
If you expect them to do something that they cannot, you might get frustrated. By understanding the differences between the two, it will help to ease that frustration.
Wildlife Acoustics Kaleidoscope software includes correlated views of the oscillogram and spectrogram. That is, the graphics are linked together according to time and share a time ruler, allowing you to easily see both views.
Within the pro version of the software, you can either rely on your own oscillogram or spectrogram analysis or use more advanced features for automated analysis – letting your computer do the work for you.
For example, Kaleidoscope Pro’s Cluster Analysis uses sophisticated pattern-recognition algorithms that scan wildlife audio recordings for animal vocalizations. Once detected, the sounds are sorted into groups of similar sounds called “clusters.” From there, clusters can be labeled for species inventory or annotated to create classifiers that can be run on additional recordings.
Additionally, for bat researchers, a bat auto-ID feature in Kaleidoscope Pro analyzes recordings of bat echolocations and automatically suggests the most likely bat species. It also provides an efficient workflow for manually vetting these auto-ID identifications, reducing the amount of time you spend analyzing your recordings.
Ready to get started? Learn more about Kaleidoscope software now.
Wildlife Acoustics, Inc.
3 Mill and Main Place, Suite 210
Maynard, MA 01754-2657 USA
+1 (978) 369-5225
+1 (888) 733-0200