Article View: adobe.audition.windows
Article #2892Re: Dynamic Processing \ Attach Release
From: SteveG
Date: Sat, 23 Jul 2005 22:21
Date: Sat, 23 Jul 2005 22:21
9 lines
2796 bytes
2796 bytes
I'm assuming that you mean attack/release - and no, that's not at all what you should be using. If you want to process the dynamics of a speaker, you need to learn what the processor does. Start by normalising your file to 0dB (all Dynamics Processor settings are related to 0dB as a peak level). Normalising moves the whole file level up so that the peaks are in a fixed place, but doesn't alter the dynamics at all. Now, on the front page of the Dynamics Processor (the 'graphic' tab), select one of the Hard Limit presets, (try 9 or 12dB) and note what the effect on your file is when you apply it. What's happening is that the 'highest' (loudest) parts of the waveform are being reduced in level, whilst the quieter parts are remaining at the same level, and that's what you need to do with your speaker's voice. That's also what the 'tipped over' part of the diagonal line is telling you - the point at which things won't get any louder. Strictly speaking, this is limiting, not compression, but it works quite well in these circumstances. You may, when you find a suitable setting, want to explore the options with gain attached to them - these are the ones that say w/boost at the end of the preset name - they will automatically boost the output back up to about 0dB again. Technically, this is called 'make-up gain', and is what you would otherwise have to do manually by renormalising after you've compressed your signal. All that the attack/release control does is to affect the speed at which compression initially takes place. With speech, you need a short attack, and a longer release. By and large though, the presets have this set about correct for speech. I'm not a huge fan of using presets, but for speech use, you can do quite a bit with the hard limit ones - as long as you exercise a little care, and don't overdo it. You'll know when you've done that, because the background will start to sound very loud in comparison to the speech, which will sound somewhat lifeless - you really shouldn't aim to reduce all of the dynamics too dramatically, because this will detract from the message somewhat. I hope that's enough to at least get you started, anyway.
Technical Details
Message-ID:
<3bbb7475.0@webx.la2eafNXanI>Path:
arc1-ca.pugleaf.net!g2news1.google.com!news4.google.com!news.glorb.com!nntp.waia.asn.au!202.72.130.18.MISMATCH!quokka.wn.com.au!webxReferences:
<3bbb7475.-1@webx.la2eafNXanI>Show Full Headers
Complete Article Headers
X-Google-Thread: 1f0032,f10228344379d6e8 X-Google-Attributes: gid1f0032,public X-Google-Language: ENGLISH,ASCII-7-bit From: SteveG <member@adobeforums.com> Newsgroups: adobe.audition.windows Subject: Re: Dynamic Processing \ Attach Release Message-ID: <3bbb7475.0@webx.la2eafNXanI> Date: Sat, 23 Jul 2005 15:21:43 -0700 References: <3bbb7475.-1@webx.la2eafNXanI> Lines: 37 MIME-Version: 1.0 Content-Type: text/plain; charset="ISO-8859-1" Content-Transfer-Encoding: 8bit Path: arc1-ca.pugleaf.net!g2news1.google.com!news4.google.com!news.glorb.com!nntp.waia.asn.au!202.72.130.18.MISMATCH!quokka.wn.com.au!webx