Some questions regarding ''DC offset'' and Normalization -Sound Forge - PC

Subject:Some questions regarding ''DC offset'' and Normalization Posted by: WJW Date:2/12/2001 12:57:27 PM I've been told that one of the first things that should be done in processing a file is to remove any DC offset adjustment. True? I'm dealing with files that are lectures- all speech. I've been Normalizing using the -10dB (speech) setting, and peak value to 0dB if clipping occurs. But I've been told that I should never normalize to 0dB. True? So far, I've been leaving all other parameters in the 'preset' modes as they appear in the Normalize 'window'. Any suggestions as to other settings to try for: Attack and Release times, or Scan Setting? Should I try Normalizing using Peak Level, or the RMS presets. I've been 'checking' the Equal Loudness Contour box. Good idea? I'm not that worried about clipping, as the files are recorded to -6dB as the highest peak. Any help would be greatly appreciated, WJW
Subject:RE: Some questions regarding ''DC offset'' and Normalization Reply by: garrigus Date:3/3/2001 11:15:21 AM Hi WJW, Yes, you should always remove any dc offset before processing your audio data, otherwise it can affect the quality. And yes, you should usually normalize to a level just below 0dB, rather than exactly 0dB. This is especially true if you will be converting your audio to a compressed format such as RealAudio or MP3. Some converters don't handle 0dB levels too well, and they need a bit of headroom to work. Best, Scott -- Scott R. Garrigus - Author of the Cakewalk Power! and Sound Forge Power! books, and Publisher of the DigiFreq music technology newsletter. Learn about cool tips and techniques for your music software by getting a FREE subscription to DigiFreq... surf to: http://www.garrigus.com/

Subject:Some questions regarding ''DC offset'' and Normalization

Posted by: WJW

Date:2/12/2001 12:57:27 PM

I've been told that one of the first things that should be done in processing a file is to remove any
DC offset adjustment. True?
I'm dealing with files that are lectures- all speech. I've been Normalizing using the -10dB (speech)
setting, and peak value to 0dB if clipping occurs. But I've been told that I should never normalize to
0dB. True?
So far, I've been leaving all other parameters in the 'preset' modes as they appear in the Normalize
'window'.
Any suggestions as to other settings to try for: Attack and Release times, or Scan Setting? Should
I try Normalizing using Peak Level, or the RMS presets.
I've been 'checking' the Equal Loudness Contour box. Good idea?
I'm not that worried about clipping, as the files are recorded to -6dB as the highest peak.
Any help would be greatly appreciated,
WJW

Subject:RE: Some questions regarding ''DC offset'' and Normalization

Reply by: garrigus

Date:3/3/2001 11:15:21 AM

Hi WJW,

Yes, you should always remove any dc offset before
processing your audio data, otherwise it can affect the
quality.

And yes, you should usually normalize to a level just below
0dB, rather than exactly 0dB. This is especially true if
you will be converting your audio to a compressed format
such as RealAudio or MP3. Some converters don't handle 0dB
levels too well, and they need a bit of headroom to work.

Best,
Scott

--
Scott R. Garrigus - Author of the Cakewalk Power! and Sound
Forge Power! books, and Publisher of the DigiFreq music
technology newsletter. Learn about cool tips and techniques
for your music software by getting a FREE subscription to
DigiFreq... surf to:
http://www.garrigus.com/