[Top] [All Lists]

Re: Secondary Harmonics

Subject: Re: Secondary Harmonics
From: "Raimund Specht" <>
Date: Tue, 03 Jun 2003 23:06:23 -0000
--- In   wrote:
> Dear Experts,
> I seem to remember form my class at Cornell many years ago that 
> harmonics as shown on a spectrogram are really a result of the 
> equipment and filtering these out of a cut will not degrade the 
final audio product. 
> Actually, what I am trying to determine is if I filter a cut to 
reduce insect 
> noise and this noise is also in the band where the secondary 
harmonic is, will 
> I be significantly degrading the final audio product?

Any non-linear element in the signal path of the recording equipment 
may produce spurious (secondary) harmonics. The parameter 
quantifying these effects is the distortion (THD) specified in 
percent. This distortion can take place within the microphone 
(especially when we are close to the maximum sound level specified 
by the manufacturer), in the pre-amplifier (at higher gain settings) 
and of course, any clipping would be an extreme source of distortion 
when the recording level was not adjusted carefully. A limiter would 
also add false harmonics. Analog tape recorders may introduce 
additional distortions caused by the properties of the magnetic 
tape. However, when using high-quality microphones with flat 
frequency responses and digital recording gear, the distortion added 
by the equipment should be very low and usually not visible on a 
spectrogram. Things might get more complicated when using a 
microphone that lacks a flat frequency response. A parabola may add 
more gain to the harmonics than to the fundamental. In that case, 
some kind of low-pass filtering would be appropriate in order to get 
a flatter overall frequency response.

Most animal sounds, even those referred to as pure-tone whistles, 
actually have some weak primary harmonics. Only a perfect sine 
generator (or a less perfect generator in conjunction with a steep 
low-pass filter) were able to produce whistles without any 
harmonics. But such ideal sine generators are difficult to find in 
animal vocalizations...

Generally, all harmonics will influence the "color" of a sound (we 
can hear that when listening to the same note played on various 
musical instruments). It depends on the intensity of these harmonics 
and their pitch whether they are important to our perception or not. 
Listening before and after filtering would reveal their importance. 
In many whistle-like bird songs, the harmonics are not very 
important and filtering them out would not significantly degrade the 
final audio product. However, applying a low-pass filter to a 
recording is similar to reducing the sample rate and may therefore 
lead to some degradation. It depends on the personal preferences, 
whether to remove insect noise at the expense of loosing some other 
high-frequency signals.

Raimund Specht

Avisoft Bioacoustics


<Prev in Thread] Current Thread [Next in Thread>

The University of NSW School of Computer and Engineering takes no responsibility for the contents of this archive. It is purely a compilation of material sent by many people to the naturerecordists mailing list. It has not been checked for accuracy nor its content verified in any way. If you wish to get material removed from the archive or have other queries about the archive e-mail Andrew Taylor at this address: andrewt@cse.unsw.EDU.AU