Configuring the Microphones Using G-Ware

Remember that each preset has its own microphone settings. It is also likely that the microphone settings in one preset may not be completely appropriate for another preset. Testing is very important in this area in order to determine optimal audio settings.

As you configure the microphones, monitor the audio with another machine so you are able to get "real-time" audio feedback. Running the RTP Packet Reflector would still be preferable to doing an audio test with a remote site. Finally, you will still be able to detect echo even though by using the RTP Packet Reflector you are in a "loopback"/echo mode already.

It is recommended that you consult the user manual in order to have more detailed information for audio terms and setting explanations beyond what is described here.

Described here is the graphical interface for inputs 1-8. The other graphical interfaces for other inputs and outputs are somewhat similar in structure, but consult the manual or the G-Ware online help for more information on these other graphical interfaces. To display the graphical interface for inputs 1-8, click the Inputs 1-8 button on the matrix or the flow-screen.

Then the following window appears:

There are a few different sections of importance here with duplication of controls and interface elements for each microphone. The first portion near the top of this window contains these buttons:

• Mute - Toggles the microphone input on/off. When off, the bar on the button will highlight red.

  • Unused mics will add "space" or a "canny", "bathroom" type of sound to the overall mix as they pickup more reflected sound from the walls than direct sound from a participant. If it is obvious that a mic will not be used for a meeting, mute it to improve the overall quality of sound.

• AGC/SL - The Automatic Gain Control attempts to maintain the audio signal at a target level. If the level goes above the threshold, the audio signal is attenuated, if the level falls below the target the audio signal is boosted. However, due the variance in level when you have someone close to the mic and another person farther away, this boost will also raise the level of the ambient noise in the room. The result is a "jumpy","hicup" sound that makes the material difficult to understand. To avoid this, adjust the Response Time to an appropriate amount or leave this option disabled.

• P Pwr - Phantom Power is needed by some microphones, check the owner's manual for the mic you intend to use.

• NC - The Noise Canceller attentuates the audio signal by the amount specified when there is a low level signal present.

• AEC - Acoustic Echo Cancellation monitors the signals fed to it and removes those signals from another signal as configured in the matrix.

• Filters - There are four parametric filters available on each channel. These boost/attenuate certain frequencies to adjust for the acoustics of the room.

• Gate - The gate parameters control whether an audio signal is allowed to pass through the channel based on the level present in the signal.

The second part from the top labelled "Gain Coarse/Fine" controls the gain of the microphone. The coarse gain can be used to adjust for various mics, or line input if set to 0db. The fine gain can be used to adjust the level for your participants.

The third part of the interface labelled "Meters" will display the gain output of the microphone at various points in the audio channel. Use the chart at the top to determine the appropriate point to monitor with the meters. Using the meters together can give you a visual idea of what the various settings are doing to the signal.

The next few pages will examine some of these settings in more detail.

A note about distortion

There are two types of distortion, analog and digital. Analog distortion is caused by sending too powerful a signal into the audio system. This sound can be associated with the sound of electric guitars in most rock 'n' roll music. When acoustic distortion occurs in a voice only signal, the content can be hard to understand. You can listen to an example here . When recording analog audio the signal is measured along a continuum from some negative number to some positive number, usually -90db-+10db. With 0 representing unity, this means that the audio level going into the system stays at the same level going out of the system. This is the ideal for analog systems in order to maintain the highest signal to noise ration. As the level goes above unity, some degradation to the signal occurs. If the signal goes too high, significant distortion can occur and interfere with the intelligibility of the signal.

When recording audio digitally, the signal is measured as negative numbers with the maximum level being 0. Any signal that goes beyond 0 is literally "chopped off" (see pic below). When this happens, one can imagine that the speaker diaphragm is suddenly stretched to its limit causing a snap like sound. If the signal remains above 0 the speaker is suddenly streched to the opposite limit of its range of motion and this causes another snap. The resulting audio from a signal such as this is unintelligible and there is no way to correct the problem. You can listen to an example here . Listen to the "one", "five" and "seven". The distortion is clearly evident. Often this type of distortion is caused when a mic level is too high or the participant is too close to the mic.