Running On Auto

No one mixes the music for that room. No one adjusts volume or tone to suit the desires of that audience. No one waits to push up a fader when the next person speaks. It’s automatic! In fact, the ubiquity of sound systems can be attributed to the large number of sound systems that run unattended.

There are many audio applications that require sophisticated mixing techniques, comprehension of complex musical relationships or the ability to make changes with only visual cues. However, there are other applications that are best suited to the automatic microphone mixer. It doesn’t become bored and forget to turn on the chairman’s mic; and it can stay attentive right to the end of an week-long seminar on accounting. This makes the automixer far more suited than a manual audio mixer in a wide range of audio systems. Why waste your incredible audio talents on something a machine can do?

Before Dan Dugan brought his practical ideas to the task of managing multiple microphones, automatic mixers were clunky devices. They simply gated a microphone on when threshold levels were reached and slammed them off when the voice level fell below the threshold. Dan designed a mixer that democratically allowed all the microphones to share the maximum amount of gain available. This is known as NOM (Number of Open Mics) automation, and allows the setup to simply adjust for the maximum level (often a reasonable margin below the onset of feedback) the system can reach. The mixer then turns on microphones that rise above a preset threshold (active) and when more than one microphone is triggered on, the output gain is adjusted to keep a constant level (reduced 3 dB with each doubling of open microphones). This would be the same way a cautious technician might leave all the faders down a little bit and then bring up the active talker’s mic to unity. This way, microphones do not need to be shut off and then suddenly pop on part way through the first syllable; like a inattentive mixing operator using the mute buttons to enable microphones.

Automated mixers offered many additional parameters, including the level of the off-attenuation (how far below unity gain that virtual fader rests); the rate of change allowed when turning the mic channels on and off; automatic level control to reduce a nervous talker’s movements influence on their reinforced volume; how many microphones would be allowed to be on at the same time; which microphones have greater priority; and leaving the last microphone on so that recordings don’t sound oddly silent between periods of talking. An important feature, which should never be overlooked in automixers, is the ability to determine a common voice appearing in several microphones. This may happen when someone moves between mics setup on a table; or when a talker wears a wireless lapel mic and walks up to a lectern mic. All good mixers (automatic or human) should be aware of this and not allow two mics to pickup one voice. Not all automixers get this feature right.

Of course, automated mixing has taken advantage of DSP to offer a whole new range of possibilities. Digital signal processing was used to reduce the echo of voices on the far end of a conference call. When designed in conjunction with the room acoustics, even large groups of people can sound like individual callers on a telephone. This was the first major stride forward in using this new processing power. Now there are more features being added to these mixers. Some are familiar to everyone that’s used a DSP box in the past decade, such as parametric EQ, matrix mixing, remote control, recalling preset configurations, compression, and sometimes even signal delay. In addition, the control over features that were common in analogue predecessors can be extended to allow much greater customization.

The most lauded recent innovation in automixers is the reduction of ambient noise. This feature is unlike the old-fashioned noise gate that merely closed unused channels. Instead, it uses a noise reduction algorithm familiar to those with audio software restoration tools, where an FFT-based filter is developed that matches the quality of the background noise, without affecting the foreground sound. The cell phone-like digital artifacts of excessive use are also familiar to those that have used the software restoration tools. With optimal settings (usually between 6 dB and 10 dB) the voice quality is maintained but dramatic reductions in background noise are achieved. This is especially important when multiple microphones are on.

How well the sound is automatically adjusted depends on the budget, design and configuration resources that were applied to the sound system. Even the most sophisticated automixing will not change the need to carefully consider and test the configuration to ensure the automated features suit the application and the overall system.

United Entertainment Media Inc.
460 Park Avenue South, 9th Floor
New York, NY, 10016
Ph. 212-378-0400
FAX 212-378-2160
by e-mail

page construction