It's funny- on the most-used local 2m repeater, we don't have a problem with the majority of users, whose voice levels are fine. I think compression is necessary to cut back on a few users who 1) practically shout into the mike, or 2) have bumped up the deviation on their radios because they think that "more deviation is better", or 3) are using a multiband radio with modulation still set as for HF, and don't realize that their excessive deviation is causing distortion in the repeater. In any case, it's a few users who are too loud, rather than a few users who are too soft.
As a road-show sound engineer and recording-studio operator in an earlier life, I know the benefits of seamless compression. The primary rule is that the static level should have no gain or compression at all, so that the compression begins only when audio exceeds a certain level. An audio compressor is misapplied if it always brings up the noise level between words. My primary audio treatment device was a DBX 166, which also has a noise gate. The trick to using a noise gate properly is to set it so that it opens at the beginning of the first syllable. I spent a lot of time getting the levels and timing fine-tuned so that the compression and gating were undetectable. It can be done with quality equipment, but good audio processing equipment is not cheap. I have found that it is helpful to simply advise a user that he is too loud and needs to back away from the mike, or that he is barely audible and needs to speak up! 73, Eric Lemmon WB6FLY -----Original Message----- From: [email protected] [mailto:[email protected]] On Behalf Of Paul Plack Sent: Sunday, August 09, 2009 10:10 PM To: [email protected] Subject: Re: [Repeater-Builder] An advocate for a little audio compression Skipp, I generally agree, but it's not the fault of the user's voice. It's a lack of training in mic technique, sometimes combined with audio circuits that aren't easily user-accessible. Compression on the repeater eliminate's the user's need to get things right at the source, and one day, he's going to need to operate simplex. I've worked with broadcast compressors for many years, and agree they could play a useful role in repeater audio chains. But I always wanted to design one that was a little different, and digital control of an analog signal path seems like a good candidate. Specifically, I'd like to have something like a compressor with very fast attack and infinitely long release, immediately dropping gain as needed to accommodate voice peaks, but not releasing until COS dropped. This would essentially set the audio gain individually for each user at the start of a transmission, without any ongoing compression to create the obnoxious "pumping" artifact we all know and hate. The downsides would be additional background noise before the first syllable, and difficulty in distinguishing users with low audio from users with inadequate signal strength. Both would feature increased background noise as a symptom. Then again, IRLP users hand out S-meter reports from a thousand miles away, so maybe it doesn't matter...(sigh) Just running the audio gain 6-10 dB hotter into a fast limiter still allows great disparity in perceived loudness, but at least the guys with low audio can be heard. 73, Paul, AE4KR ----- Original Message ----- From: skipp025 <mailto:[email protected]> To: [email protected] <mailto:[email protected]> Sent: Sunday, August 09, 2009 9:07 PM Subject: [Repeater-Builder] An advocate for a little audio compression ...a number of operators don't seem to have voices that drive their radios with adequate audio...Consider 6 to 10dB of audio compression in your repeater system... . <http://geo.yahoo.com/serv?s=97359714/grpId=104168/grpspId=1705063108/msgId= 93195/stime=1249873641/nc1=4025373/nc2=5689702/nc3=5807838>
