DNA1 De-noiser De-clicker Ambience Processor
The Weiss DNA1 is an extremely versatile toolbox for restoration work. It handles broadband noise reduction as well as de-clicking and de-crackling. The K-Stereo Ambience Processor brings back life to the music in case it lost some during de-noising, or generates a pseudo stereo signal out of mono. Finally an M/S matrix allows to define the stereo width.
DNA1 – a milestone in the Gambit Series equipment developments at Weiss Engineering. Three state of the art algorithms are packed into the DNA1. The de-noising section has been developed in cooperation with the Pure Notes Company (UK), the de-crackling and de-clicking algorithms are the Weiss R&D lab’s design and finally the K-Stereo Ambience Processor we licensed from Mr. Bob Katz (USA). All functions can be used concurrently. The DNA1 can be switched to M/S mode which can be useful e.g. for de-noising the M channel only while applying less de-noising on the S channel to leave the ambience intact.
There are a total of 15 knobs in order to almost have a “one knob per parameter” operation. The knobs are touch sensitive to switch the LCD to display the associated parameters. The LCD shows an FFT of the input signal and meters to indicate the amount of de-noising going on. In addition all parameters are shown in numerical format.
Snapshots, MIDI, metering
Other basic features are an A/B compare memory, a 128 position snapshot bank with two additional banks for back-up (can also be dumped/recovered via MIDI), over indicators, MIDI control for each parameter, an overall bypass switch and an overall gain control, a channel ganging switch and an M/S mode switch.
Signal processing, connectivity
The internal processing is done in a 40 Bit floating-point format. Input signals can be between 44.1 and 96 kHz. The output can be POW-R dithered to 16, 20 or 24 Bits. The digital input/output are in AES/EBU format on XLR connectors.
The broadband de-noiser section works based on an FFT approach with a selectable number of frequency bands. The energy in each band is analyzed and thus determines whether the band is of relevance for the music signal. The de-noiser does not need to take a noise fingerprint, it continually adapts to the noise signature. Three operating modes can be selected, the automatic mode for uncritical de-noising, i.e. for cases where the noise floor is fairly low already and thus the de-noising is not that intrusive. The semi-auto mode where some of the de-noiser’s parameters are estimated first based on the auto-mode and subsequently are refined manually by the user. And finally the fully manual mode where all parameters are accessible, for those cases which require more subtle tuning to get the best result. Like e.g. cases of old recordings with high levels of noise which need a subtle balance between removed noise and noise required to stay such that the “vintage” touch of the recording is kept alive.
The de-clicker section is made up of nine main parameters. There are three groups of three parameters each. Each group, named “quality”, “sensitivity” and “click type”, has a few individually selectable snapshots which allow to zoom in on a particular de-clicking problem at hand. After setting the three snapshots for the three groups, the parameters can be trimmed further individually. With that scheme it is relatively simple to get a result quickly.
The K-Stereo processor uses one main parameter, the K-level, which defines the level of the effect. In addition there is a three band equalizer with low cut, high cut and a parametric center band in peaking mode. The equalizer shapes the audio going through the K-Stereo process. The “wide” and “deep” on/off switches allow for additional shaping of the effect.
Output M/S Matrix
The output section consists of an additional M/S encoding/decoding matrix for final stereo width shaping. In addition left and right channels can be trimmed individually plus finally the volume of the stereo signal can be changed with a single knob.