Burned Audio CD diferent volume.

[GPo]

Anyone know how to set up an audio cd so that the songs do not differ in volume.
I frequently find myself adjusting the volume up or down between songs of my burned audio CD's.
I use adaptech easy CD
GPo

[daveleau]

This is a difficult task as each cd is mastered differently. There is a program that will do this for you. I cannot recall the name but I will find it and let you know.
Dave

[jad1097]

You need a program that will normalize the songs before burning them. I use MP3 Cd Maker.

[Solidus]

The reason may be that the mp3 files have a different bitrate. I think that may cause some to be louder.

[gmcpher196]

Use the same bitrate.

[jim.t]

I use Goldwave sound editor. Most of the sound normalizers set sound limits with all tracks, but it may appear to be louder on some tracks due to equalization/deeper bass-treble/etc. When I assemble tracks for burning, I check their sound and equalize them also so that they appear to have the same sound level.

[GPo]

Thanks people.
GPo

[eagle1]

Hey GPo! I asked the same questions a couple of weeks ago because I was having the same problem and instead of downloading new software to "normalize" the files, I found out the Nero Burning Rom (the cd-mastering soft that I use) has that option integrated. I have Easy CD Creator also and I think the software was this option also.

Just select the files and adjust their properties by right clicking the files (right click = menu + properties). There should be an option that allows you to normalize the songs..!

[crazyray]

The most common reason the level varies on MP3 files is the source CD's! Current pop CD's are invariably compressed and boosted up to 100% peaks. Many older CD's may have a much lower level. Normalizing can help keep you from reaching for the volume control with every new track on compilation CD's. I use the one in Audiograbber.

[Joel Kleppinger]

crazyray is quite correct as to the cause (although 100% (I'm assuming infinity:1) compression isn't exactly correct... the compression is usually less... toward 2:1 to 5:1, depending). Now those extra loud car commercials... that's as close to infinity:1 as they can get. http://sysopt.earthweb.com/forum/smile.gif

Many of you might think of compression in terms of zip files rather than sound compression. Sound compression is acheived by specifying a cutoff volume (like say -18db with 0db being the point at which clipping occurs) and a ratio for compression (say 2:1). Other variables, such as Attack and Release times, can be adjusted, but this is irrelevant to the discussion. All sound that cross the threshold undergoes compression as specified by the ratio. So, with -18db threshold at a 2:1 compression ratio, a spike that hits -6db (12db above the threshold) would be compressed to -12db (6 above). After a track has been compressed, it can then be gained up much further, making the whole thing sound louder (although the loudest sounds are somewhat mixed more into the rest of the mix).

Bitrate is basically irrelevant to the actual volume (especially at 128kbps and above).

[Tom Teriffic]

Just to add another nickel's worth here....

Normalizing is a function of taking the absolute loudest portion of a given audio file, and running it up to just under "digital zero" or the point at which digital distortion sets in. This then sets the level for the entire song. Mostly, it'll work pretty well. The trouble can arise with the previously discussed compression, amounts of which can vary wildly, from none at all on an acoustic or classical piece, to the supersaturated multiband compression used by FM radio in an attempt to be the loudest signal on the dial.

If the tunes in question are commercial releases, odds are good that simply normalizing thenm can get you pretty close to an even volume, allowing for different emphasis in various frequency bands (bass, midrange, treble) that any given tune might have.

However, if you're still confronted with a piece or two which doesn't seem to be up to par, level-wise, try this:

Most audio waveform editors (and Adaptec Easy CD Creator "Deluxe" has one) will allow you to zoom in or out on various portions of your sound file, which is represented by a trace which looks somewhat like an oscilloscope trace.

Look at the song in its entirety. If most of that squiggly trace, or energy, seems to be occupying a relatively narrow area across the center of the window, punctuated by one, a few, or even a couple dozen "spikes" which go all the way up or down to the limits, then:

1) Select one of the spiky areas.

2) Zoom in and in and in until you are seeing just the offending spike, or musical peak.

3) Techniques will vary from editor to editor, but most will give you an option to "redraw" that waveform, or to grab a little "handle" on the waveform with your mouse, thereby allowing you to drag the peak up or down.

4) Redraw or "redarag", if you will, the peak downward (in the case of an upward-going peak. For a downward-going peak, do the opposite, obviously), taking care to recreate the original shape of the peak, just at less of a level level.

5) Repeat as necessary.

This can be a very handy tool. I've used it on several original productions when I just couldn't get an overall level that sat well with other cuts. The alternative was some serious compression known as limiting, which brings a whole 'nother set of audible artifacts with it.

I believe Goldwave gives you this option, and there's an excellent inexpensive audio editor out there called Cool Edit 2000, available at:
<A HREF="http://www.syntrillium.com" TARGET=_blank>http://www.syntrillium.com</A>

It offers a whole gang of tools with it and might make a good choice for this type of work.

Okay, it was more than a nickel. Sorry :-)

Tom

[This message has been edited by Tom Teriffic (edited 01-17-2001).]

[This message has been edited by Tom Teriffic (edited 01-17-2001).]