Digital Signal Processing
Because an acoustically designed and constructed listening room is an expensive build, especially in
controlling low frequencies. Most of us are constrained by our existing rooms that we listen to music and
movies in. DSP affords us a way to mitigate these low frequency issues for a fraction of the cost of a new
room build, but gain the benefits.
The two most important aspects of a good sounding bass in a
room are smooth low frequency response and reduced low frequency room reflections.
is even sounding bass that is crystal clear.
This book provides the audio enthusiast with an easy-to-follow step-by-step guide for designing a custom
digital filter that corrects the frequency and timing response of your loudspeakers in your listening
environment so that the music arriving at your ears matches as closely as possible to the content on the
Industry guidelines, spanning over 40 years of evolution, are referenced throughout
the book, providing the recommended target responses for accurate sound reproduction. Correcting the
measured response to known target responses is, in effect, matching the acoustic output signal arriving at
your ears to the digital audio stored in a media file. The target responses can also be tailored to your own
"For anyone with an interest in DSP technology and how it can be applied to improve the performance
of real loudspeakers in real rooms, this is a fascinating and very affordable e-book. It is both
well-written and comprehensively illustrated, and guides the reader step-by-step through the concepts,
techniques, and practical implementations of both the science and art of loudspeaker correction."
Hugh Robjohns, SOS book review.
Reproduction Using DSP
After DSP, the low frequency response is calibrated to an industry standard operational room response
curve within ±3 dB tolerance. The result is even sounding, smooth bass response, at multiple listening
positions. By calibrating to an industry standard operational room response, we get the benefit of hearing
the same tonal response as the artists intended.
Click to Enlarge
There is a 20 dB sound pressure level (SPL) variation at low frequencies in the before frequency response.
This is simply due to the physical dimensions of ones listening room. Virtually all small room acoustics
that are our living rooms and home theaters will have this issue of large, uneven bass response
To our ears, a 20 dB SPL variation is perceived to sound 4 times as loud or quiet
depending on which frequency range is being compared. The bass will sound boomy on some notes (i.e.
frequencies) and all but disappear on other notes.
Frequency response calibration
Transient Response Analysis
Given that the envelope of the dips at low frequencies are based on room modes, which are standing waves, we
know at different mic positions we will measure somewhere on the "up and down" of each wave (i.e.
frequency). By correcting only the envelope of the wave, or "average" over time, is how a smooth low
frequency response over a broad listening area is obtained with two measurements at one mic location.
high frequencies the envelope sits high in the "comb filter," which avoids over correction of the
dips. This is key for natural sounding high frequencies, where in combination with Frequency Dependent
Windowing (FDW), described below, provides broadband "tonal" correction of the direct sound above
the room's transition frequency without over correcting the dips.
The science behind this,
and FDW described below, is from James (JJ) Johnston's, "Acoustic and Psychoacoustic Issues in Room
Correction." plus PowerPoint presentation. JJ explains why we hear what we hear in small room acoustics.
Our ears perceive that music is mostly transient in nature. The DSP software we use analyzes the full
frequency range of the transient behavior and this analysis leads to a new frequency response that follows
the peaks, but not the dips as can be seen in the frequency response chart above.
If one studies
the chart closely, one can see the envelope response is following the peaks but not the dips. Note the
marker at just over 100 Hz., one can see a very narrow dip, but the new analysis does not follow that narrow
dip. This new frequency response can be considered as an upper envelop of the original magnitudes. The
spectral envelope is used as the basis for further calculations.
Click to Enlarge
Frequency Dependent Windowing
Before the explanation, we do not apply in-room equalization to achieve a flat steady-state frequency
response at the listening position. This is not the desired target response, nor is this how we interpret
the industry standards operational room response curves.
We use DSP to calibrate sound systems
below 600 Hz to smooth out low frequency response and remove room resonances. A well designed loudspeaker
will have a steady-state in-room frequency response curve that is tilting downwards. This is because
loudspeakers are omnidirectional at low frequencies with narrowing directivity at higher frequencies. The
rising bass energy yields a steady-state room curve with a downward tilt.
The state of the art DSP software we use applies Frequency Dependent Windowing (FDW) so that we control the
dividing line between in-room steady state response and the direct sound from the loudspeaker. This dividing
line is typically around 600 Hz. This means below 600 Hz we are taking the room into account and above 600
Hz just the direct sound from the loudspeaker. This is because below 600 Hz or the rooms transition or
Schroeder frequency, the room is in charge. Above the rooms transition frequency, the loudspeaker is in
The chart example above displays an impulse response and three window widths at different
frequencies using 15 cycles as the FDW setting. The horizontal scale units are in 100 millisecond
increments. We calculate the FDW sizes using these FDW math equations. Using the FDW math equations, a 20 Hz
window width is 750 milliseconds (ms). At 100 Hz the window width is 150 ms and at 1 kHz it is 15 ms.
While this example uses 15 cycles as a single FDW setting, the DSP software we use can adjust
both low and high frequency FDW cycle times independently. This allows us to calculate the optimum FDW sizes
based on your room's size and statistical analysis of model distribution and early reflections density.
This analysis gives us the smoothest integration between "room plus loudspeaker" below your
room's transition frequency and "loudspeaker direct sound" above your room's transition
If the loudspeaker requires a spectral tilt above 600 Hz due to the loudspeaker being
a bit too bright or dark sounding in your room, we equalize the direct sound only. We only apply broadband
“tone control” like spectral tilts to match industry standards operational room response curves.
Click to Enlarge