|
| AudioIssues |
UserPreferences |
| AG Wiki | FrontPage | RecentChanges | TitleIndex | WordIndex | SiteNavigation | HelpContents | moin.sf.net |
This page is meant for folks to list audio issues they've run into when using the access grid.
Some off the top of my head are:
One of the easiest ways to solve a lot of audio problems is to use a second "monitoring" machine with a pair of headphones to listen to the audio during a meeting. This allows the node operator to hear the audio from their node and determine how it sounds in relation to the other sites. This can also alert the node operator that their site is causing echo. To set up a monitoring machine, just install the AGTK v2.x and only set up an Audio Service. The video services are not necessary. Then navigate this monitoring node into the venue.
the main problem is there are too many controls spread throughout the system. Our system has 12-15 gain controls just to get the audio from the mics to the AG and from the AG to the loudspeakers. I see the greatest need in two areas: active echo cancellation happening in RAT rather than an outboard unit. If echo cancellation can be done in RAT, that eliminates the most complex and expensive part of the current AG system.
an upgrade to the RAT gui to make it function more like a typical audio mixer. currently the controls and meters are oriented the wrong way. Audio mixers typically use a vertical orientation for meters and level controls. Rotary controls are not practical in my experience with software that uses them. Also, a way to combine channels into subgroups and then have them split out into individual channels using a key combination would be nice. This way you can quickly isolate an offending channel and adjust them independently of other channels.
other things: The idea of an audio compressor, though implemented in the XAP series of EC devices, needs a slight redefinition. Typically a compressor turns the volume up or down (ratio) at a particular speed(attack/release) when the signal moves beyond a certain threshold. For the purposes of AG an addtional threshold needs to be defined such that the volume is boosted/attenuated to fall between the threshold points. This is slightly different than AGC which endeavors to keep the volume at a certain level. This new compressor paradigm would endeavor to keep the level within a certain range. the upper threshold would serve as the "limiter" point above which a signal is attenuated and the second threshold would define the "noise floor" below which only noise is heard. As the level rises above the limit, the level is brought down to the limit threshold at a certian attack speed, as the level falls below the floor, the signal is brought down at a certain release speed to nil.
