Voxengo TEOTE by Sound-Guy
TEOTE Automatic Spectral Balancer plug-in by Voxengo
Voxengo is the brand name for Aleksey Vaneev who has created some extremely useful free measurement plug-ins such as SPAN and the Correlometer previously reviewed in these pages (OK, these video screens). However Voxengo also has a great range of audio processors that haven’t been reviewed here, and that is about to change.
If you haven’t discovered (and used) Voxengo tools, you are missing some truly exceptional audio processors. They include equalizer plug-ins, dynamics processors, harmonic enhancement units, saturation FX, reverbs, spatial imaging, and mastering plug-ins. Aleksey has developed unique algorithms that provide excellent audio quality and some unique processing capabilities.
TEOTE default screen with no input signal
The latest tool from Voxengo is TEOTE, which stands for “That’s Easier On The Ear”, and is new take on the dynamic EQ in that it has up to 64 frequency bands and some unique response features. The user manual states that it can automatically provide gentle resonance taming, de-essing, and tilt EQ, but I found it can do a lot more. It can operate like a multiband compressor with as few as 3 bands, but as a very versatile compressor in that it can both compress and expand audio material.
What Does it Do?
When I first tried testing it with my usual approach of measuring frequency response curves, harmonics and dynamic performance, I found almost nothing to report! It didn’t seem to do anything! However, testing audio samples of music and vocals, I found impressive results! Analytically it seems to do nothing, but subjectively it can do a lot!
For example, in addition to taming resonances and de-essing duties, I found it can increase the dynamic range of “flat” audio material. One of my first music tests (a stress test since I wouldn’t plan to use it this way very often) was to see if I could improve the sound of some tracks from a famous album by a famous band that was met with much criticism when it was first released since some tracks have dynamic range (DR) lower than 6 dB. I won’t name the album, and I will admit the extreme compression was planned and the tracks sound OK when played LOUD, but streaming services turn it down a good 12-14 dB so it sounds wimpy and flat. When I cranked the TEOTE settings up on several tracks from the album, I actually heard bass guitar and kick drum much clearer than ever, and one track with its original DR at 5 dB was increased to over 12 dB of dynamic range! Songs really sounded better than I’d ever heard from this album. This rather surprised me since everyone “knows” that once music is badly crushed, you can’t get back dynamic range. I don’t claim TEOTE should be used to do this on your music (you should not crush the s. . t out of it in the first place!), but it is a testament to the processing power of Aleksey’s latest tool.
Stereo audio with very squashed dynamics expanded to over 12 dB by TEOTE
At more reasonable settings TEOTE can be very useful on individual tracks that need taming or need more dynamics, and on full mixes it can balance the spectral distribution and help sculpt the general frequency dispersion without overly smoothing individual frequencies.
How Does it Do It?
TEOTE has some unique controls and some that appear familiar, but don’t do exactly what you expect. The main control panel has two sections, Dynamics and Spectral Profile. Dynamics includes five rotary controls with FX being an intensity control for the processing (like a wet/dry control), Boost T being is a rather unique threshold setting, Base Atk and Base Rls are attack and release times, but not for all frequencies (they are for the 20 Hz frequency band only, unless you set Hi Timing to 1:1!), and Ch Link is a channel link control that adjusts how independent the processing is for the audio channels (up to 8 audio channels can be processed, though I tested only stereo).
TEOTE is unique in that a normal threshold setting is not used – it uses a weighted loudness estimation similar to LUFS and applies its multi-band gain adjustments relative to the overall momentary loudness level. There are also no ratio controls and Boost T (boost threshold) is anything but a normal threshold control. Boost T sets the level for each band relative to the estimated LUFS at which gain adjustment stops. Boost T has a button with two settings, U and L, with U returning each band to unity gain and L making them acting as limiters when gain adjustment stops. This all may sound complicated – because it is! The results are easy to hear and the spectral display shows in real time how gain is being applied in every frequency band.
Boost T can control rather TEOTE acts just as a “smart” compressor or as both a compressor and expander. At its maximum setting of zero, only compression will be applied and you will see only gain reduction with bars only dropping below the center line of the spectral display. When the dial is turned down towards -16, the range of expansion is increased and you will see more boost and gain reduction applied on the spectral display. A little confusing? Just twiddle the knobs while playing music and the spectral display will indicate what’s happening, and your ears will hear the results. You will quickly see/hear how the Boost T (and FX control) can make music clearer and more dynamic (hint: set the FX control to maximum to start and back off to taste when the effects are too strong).
Note the out/in display below the Out Gain knob – this shows the instantaneous difference between the input signal level and output, and other than with very mild settings, it will not show a steady reading but provides a way to adjust the output level to roughly match the unprocessed (input) signal level for comparison.
TEOTE processing a mix with overall spectral slope close to the targeted value of -4.5 dB/octave
Channel Link has a button that selects between Peak or Average (RMS) linking control and there are three windows, one that is labeled Mastering, one labeled Hi Timing, and there is a window above the Ch Link control with six options.
The Mastering switch will increase the precision and resolution of the processing, offering gentler gain adjustments and switching the dynamics control from feed-forward to a feedback mode. And it will eat up a lot more CPU resources! The Hi Timing adjusts how fast the highest band (20 kHz) responds with attack and release relative to the Base Atk and Base Rls settings – it is a ratio so that 1:20, as in the above image, means both attack and release at 20 kHz will be 1/20 the times set in Base Atk and Base Rls, or in this case 1.8 msec for attack and 11.4 msec for release. Bands in between will be proportionally scaled.
The “Energetic/Balanced/Controlled/Fluid/Fluid Stable/Fluid Punch” switch above Ch Link selects the overall loudness estimator’s response mode which as the manual states, affects transient response, “stability of sound”, and overall sonic coloration of the result. This control creates extremely subtle changes and I used Balanced and Energetic modes for most of my testing.
And that’s just the Dynamics section!
The Spectral Profile section has five controls that are not so familiar, although the Bands control is most obvious. It sets the number of frequency bands to be processed, from three to 64, each processed independently. As might be expected, the more bands used, the more CPU power is required. The Slope control may be a familiar concept to you since music tends to have a spectral slope, from about -3 dB/octave (which is the slope of pink noise) for much classical pop music to about -5 dB/octave or even -6 dB/octave for more dynamic rock and jazz. In TEOTE the chosen slope is a target value which the dynamics processing will attempt to match, but without eliminating characteristic peaks and dips in the spectrum, and without reducing dynamics – in fact, as described earlier, most settings will increase dynamics a dB or more.
Lo Cut and Hi Cut are what they are, providing low and high cutoff filters with a switch (Cut) to set slope to either 6 dB/oct or 12 dB/oct. These filters are most useful on specific tracks/instruments rather than full mixes. Room Dip is a special filter that applies a -2.5dB, 1-octave wide, bell-shaped dip to the profile to compensate for usual listener room’s acoustics where room modes can cause “booming” in the 100-300 Hz range, making the sound “muddy” (can be disengaged by moving it to 20 Hz). This is a common correction to make in a mix, though I usually apply it with a normal EQ.
There are two more boxes in the Spectral Profile section, the EQ type setting and the Apply to Range selection. The EQ box has three settings, Flat, Eq.Loud, and EQL+Rock. For most use like pop music, this may be set to flat – the other two settings provide bumps is several frequency ranges that can be useful in mastering rock music. I found the EQL+Rock setting to really bring up the intensity of some heavy metal songs I tested.
And I found the Apply to Range control very useful – this provides a flexible control of the frequency range to be processed – with Apply to Range turned off, the full frequency range is processed (20 Hz – 20 kHz) and the Lo/Hi Cut simply rolls off the low and high end as described earlier, but when the Apply to Range mode is active, things get very entertaining – the Lo Cut and Hi Cut prevent the dynamics processing below and above these settings respectively, so you can focus the dynamics processing to a specific frequency range. And maybe even more useful at times, if you set Lo Cut higher than Hi Cut (say Lo at 3 kHz and Hi at 500 Hz) then dynamics processing will by bypassed between these frequencies and proceed below the Hi setting and above the Lo setting! Very clever! And can be very useful on full mixes.
Graphics and Settings
Voxengo plug-ins have a standardized “look” which some people have said looks “dated”. To me they look fine and functional and I’m glad to see consistency over the different units and over the years. Some plug-ins I use take a skeuomorphic approach, looking like the classic hardware they emulate. This is not always a bad thing, but taken as a whole, digital skeuomorphism leads to inconsistency across applications, making use less intuitive than it could be. When a processor is an original design there is really no need to make it look like a 50 year old compressor, complete with front panel scrapes and chipped paint! Sound quality and usability are what really count.
Every Voxengo plug-in shares a very similar Information and Settings window to set up the look and scale of the GUI, including a selection of user interface color schemes, default auto-oversampling level, minimum or linear oversampling mode, level meter response times and a number of other user parameters – even a UI Color Editor for those who must customize.
Available in 32-bit and 64-bit AAX, AU, VST, and VST3 plug-in formats, for macOS and Windows computers. I used REAPER and Studio One for testing in a PC Audio Labs Rok Box PC (Windows 7 64 Bit, 4-Core Intel i7-4770K, 3.5 GHz, and 16 GB RAM). One instance of TEOTE uses about 60 MB of RAM. To estimate CPU load I used REAPER’s Performance Meter: the CPU requirements (processing stereo with Mastering mode off) varied with settings, mostly related to number of bands used – from about 0.1% for 3 bands to 0.5% for 24 bands to 1.2% for 64 bands (all with Channel Linking at 100%). Dropping Channel Linking to zero (full separate channel processing) increased these CPU loads to 0.15%, 0.8%, and 1.8% respectively. Note these values are with no oversampling – 4x oversampling increased CPU use by about three times and 8x oversampling by six times. And turning on the Mastering mode roughly doubles CPU use for any of these other settings. So if you want to run at extreme quality modes, you will need a powerful computer. However, audio results were excellent without using Mastering mode and I found no need for oversampling more than 2x. Latency is always zero.
TEOTE yields excellent results on many types of tracks and mixes. It can provide a range of effects from subtle tonal changes to mild dynamic control to extreme dynamics expansion, all with very low distortion and negligible artifacts. A truly fine and unique processor – check out the free trial version.
Unique audio dynamics processing system with from 3 to 64 independent frequency bands
Excellent audio quality with low distortion even using extreme dynamics processing
Very useful for both mixing duties and mastering
Includes stereo and multi-channel processing
64-bit floating point processing with up to 8x oversampling
Includes preset manager, undo/redo history, A/B comparisons and contextual hint messages
Supports all sample rates
Zero processing latency
Free trial available
The user manual is a bit short on explanations, but I suggest reading it through for a general understanding. Turning knobs, viewing the dynamic spectral display, and listening is the best way to really understand what TEOTE can do for you. There is a separate manual available (Primary User Manual) that covers all the global settings that is useful when first trying any Voxengo processors.
Moderate to very high CPU usage if you use the highest quality settings with maximum band count, but I doubt anyone really needs everything set to 11!