Dry Voice Detection - Omnia .9 Installation & User Manual

FM MENU INTERMEDIATE MODE

Dry Voice Detection

The Dry Voice Detection Menu contains the controls to enable the dry voice detector circuit and adjust the dynamics section of
Omnia.9 when this feature is engaged.
Dry voice is one of the most difficult waveforms to process cleanly, as the human voice is complex in nature and is typically
asymmetrical in form. Stations that choose to process aggressively in an effort to maximize loudness may find that bare vocals
come through with unacceptably high levels of audible distortion.
Omnia.9 overcomes this situation by automatically detecting (in the Auto Detect mode) when the input audio is dry voice
and using a separate set of multiband targets, attack rates, and release rates. This allows the dynamics section to do more of the
"heavy lifting" and reduces the amount of clipping necessary to maintain the same level of loudness.
The Dry Voice dropdown menu can be set to "Off", "Auto Detect", or "Force". "Off" completely disables this feature.
♦
"Auto" allows the processor to automatically detect the presence of dry voice. "Force" overrides the main multiband
settings and uses the Dry Voice Detection settings at all times. It is useful to force detection of voice while testing with
music, in order to facilitate adjusting the separate dry voice multiband controls so that even when the inevitable false
voice detection occurs, the frequency balance will not shift unacceptably. It could also be useful to control this parameter
by GPI or HTTP, tied to the DJ mic fader. The default setting depends upon the preset chosen; most presets have this
featured set to "Off" by default, while the default setting for more aggressive, loudness-driven presets is "Auto Detect."
The Omnia.9 processing algorithm is really optimized for music, and mostly insensitive to peaks, which is what can cause
voices to come out too loud. This is opposite of most processors, which tend to be peak sensitive, requiring pre-processing
the microphones with a dedicated mic processor for voices to come through loud and clear. Thus, many users have found that
the best results are achieved by completely bypassing the mic processing, using the existing units only as pre-amplifiers, as the
non peak-sensitive algorithms in the Omnia.9 will naturally perform the work of a mic processor, without needing dry voice
detection.
CHAPTER 11 |
63