The modern world's frenetic pace is premised on electronic highways and yet the pursuit of birds continues to bumble along on hard-wired gravel. Why? - The answer to that lies in '
... and our reliance on vision. Aided optically or otherwise, sight is unquestionably our primary sense. Fortunately both sight and hearing feature predominately in the avian world and because birds communicate by using visual and sound signals, as we do, we are able to tap into their
fairly easily.
A note of warning! - Let your hearing dance lead when sight is either obscured or it's too dark to see. Apply wonky vision, early in the perceptual process and you could fall foul of the
McGurk Effect ie:- What
we see can influence what
we 'hear'. It's a
visual-auditory cross-talk or an illusion that occurs when the auditory component of one sound is paired with the visual component of another sound, leading to the perception of a different sound altogether.
 |
| River Warbler [Locustella fluviatilis] |
The idea that birders accurately identify birds by '
overall impression / behaviour', rather than by colour, shape etc., is not only the hallmark of an A-class birder but also a focus-driven, automated response to visual stimulus. It's a skill acquired through disciplined application and focused learning rather than by osmosis through longevity. The same focus-driven learning applies in correctly identifying birds by
song or call.
 |
| A simple up-slurred song or call |
Although any exercise that helps birders focus more intently on song is beneficial, technology, fortunately, helps render subtle
song or call into a visual representation of the sound. Since vision, rather than hearing, is dominant in the perceptual process, these renditions or sonograms are a short-cut to auditory success. The sonogram or, more accurately, the
Audio Spectrogram, allows birders to
'see' a
song or call; reveals intricacies of the bird's call and exposes the student to the many subtle differences in similar-sounding
songs or calls that would otherwise go unnoticed in the field. Once familiar with the more difficult aspects of a
song or call birders can more easily focus on these features.
 |
| Down-slur with harmonics |
To make it possible to
'see' a composite sound audio is broken down into samples / segments, usually a millisecond (ms) [1/1000th of a second] in length. A more thorough investigation of band-pass filters or FFT is probably more than we need for the purposes of this discussion. Suffice to say an analyser checks each sample for sound. If audio is present the sound is analysed at each of many different frequencies to determine which frequencies are present at that point. The presence of any audio content is graphically represented by a dot at each frequency present at the time of the sample.
Sound consists of
travelling waves of alternating pressure and displacement. These waves are generated by vibration. In birds,
song or call, is generated in the
syrinx, an organ situated at the lower end of the trachea and at the junction of the two bronchi. To generate
song or call the syrinx contains membranes which are activated /vibrated by the passage of air from the air sacs.
The number of
travelling waves or '
cycles per second', is the
frequency. Frequency is measured in Hertz where 1 Hertz (Hz) = 1 '
cycle per second'.
 |
| Harmonics - 'Musical' sound |
An
Audio Spectrogram, therefore,
shows the evolution in time of sound frequency.
Frequency is represented on the vertical axis and
time on the horizontal axis. The amplitude / intensity [
'loudness'] of any frequency, at any given time / sample, is represented by a gray-scale value from white ('
loudness' = 0) to black (maximum intensity).
Interpreting an
Audio Spectrogram takes a little practice and a basic understanding of the terminology used to represent sound. TONE refers to a sound's pitch, quality and intensity. A
simple TONE or a
sine wave is the purest of all sounds and has a single frequency only. A
complex TONE is comprised of two or more simple tones or OVERTONES. The TONE of lowest frequency is the FUNDAMENTAL; the sounds stacked an octave higher than the next, with all but the lowest pitch being soft, are the OVERTONES or HARMONICS. A complex tone is much '
richer' than a simple
sine wave. The more
harmonics visible in the
Audio Spectrogram, the more MUSICAL or '
richer' the sound.
PITCH refers to the degrees of '
highness' or '
lowness' of sound or, more accurately, the
sensation of frequency. A high pitch corresponds with a high frequency whereas a low pitch = a low frequency. Expect six or seven basic pitch-patterns on an Audio Spectrogram:
- Monotone - sounds don't change in frequency and are represented as a horizontal line.
- A 'rising' pitch or Up-slur - represented by a line tilted upwards.
- A 'falling' pitch or Down-slur - the line tilts down.
- A rise followed by a fall in pitch is called an Over-slur.
- A fall followed by a rise in pitch is an Under-slur.
- A 'Trill' represents a pitch that varies quickly over both frequency and time.
- 'Noise' is the simultaneous sounding of equal levels of pitch.
Let's park the theoretical aspects of
Audio Spectrograms for the minute and look at some practical examples. The first example was a personal experience and since it pertains to my own experience we'll start there.
River Warbler [
Locustella fluviatilis] is, in the Southern Hemisphere at least, a prime candidate for
skulker of the year and the reason it rates high-up on that list is its recluse-like nature. Generally '
silent', this furtive jester of the underworld is particularly merry in dankest, darkest Africa, a murk boasted in our wet-season.
One month a year, in the austral summer, this apparition flaunts its brown splendour by serenading the gals from a few feet higher than its traditional tangled playground and it is precisely during this March-month of festivities when birders swoop on the songsters to score a glimpse, however fleeting.
 |
| River Warbler - in hand |
Perversely most River Warblers arrive well before March and go about their business unannounced. How then, in the absence of their oft-recorded
March-only trill, are they confirmed
in situ well before they themselves confirm their arrival months later? The answer to that is easily explained by
Audio Spectrogram. Like any recluse these birds value their solitude and when their harmony is disturbed they do exactly as expected, voice their displeasure. Fortunately their displeasure lacks much imagination and is replicated throughout the community. Spurned by most field guides this alarm-call / contact-call is repetitive, recognisable and one that birders can & should learn. The most complete way to do that is by
Audio Spectrogram. ie: '
See' the call, recognise the subtle intricacies, take the knowledge to the field and viola, score a '
River' and you'll do so way before the
riff-raff claim the same in March...
 |
| Expanded Audio Spectrogram - courtesy Dewald Swanepoel |
The second example is more recent and confirmed so many of my own suspicions that the lesson itself is too important to dismiss. On this point I'm indebted to Dewald Swanepoel who supplied both the sonograms and a lucid explanation for the material in this example.
The sequence of events unfolded as follows:
- An unfamiliar song was recorded before sun-up.
- The indistinct recording was sent to a local professional for identification.
- The song, as a stand-alone, was one of two species but based on the circumstances of the recording ie: habitat and time of call; the correct species was excluded in favour of the incorrect species. [Recall the McGurk Effect?]
- The song in question was reduced to Audio Spectrogram, compared with filed data, and the species-responsible correctly confirmed.
The fact that the recorded song caused some consternation / confusion, even among the very best, is testament to the subtle intricacies of
song or call across similar and even wholly unrelated species.
Reduced to
Audio Spectrogram the 'mystery' songster was quite clearly
Species A and not
Species B; a visual confirmation of an auditory debate, sans any external visual stimulus / interference, other than from the graphical rendition of the vocalisation itself.
 |
| Courtesy Dewald Swanepoel |
Dewald describes '
Species B' in his expanded Audio Spectrogram as follows:
'.. each phrase consists of 4 introductory notes, the first group of two being similar (high-low-high between 5.7 kHz and 4.4 kHz) and spaced 430 ms apart, the second group of two being similar (high-low between 4.8 kHz and 3.7 kHz) and spaced 240 ms apart as well as following 240 ms after the start of the second note in the first group of 2.
After these, the bird breaks into the trill which consists of these compound high-low-high-low sweeps at a very consistent rate of just ... over 10 notes per second. Each note consists of a high-low-high-low sweep from 6 kHz to 3.5 kHz, back up to 5.3 kHz and down to 2.5 kHz. Each compound sweep happens in the space of 50 ms with another 50 ms pause before the next compound sweep.'
 |
Barratt's Warbler [Bradypterus barratti]
|
Dewald's understanding of the intricacies of sound, represented in an
Audio Spectrogram, is advanced and we are indebted to him. Notwithstanding, even an unpracticed eye can '
see', by simple observation and at a glance, that the 'Mystery' bird is '
Species A' rather than 'Species B'.
If our impulse to use tools is the product of natural selection, then there must exist a feedback loop between technology and survival .. The only question that still begs an answer then is this -
- Is '
Traditional Field-craft' just
organised technology resistance?
No comments:
Post a Comment