I was refered here by a guy from videohelp.com and he said that you guys were "They're sound scholars." and that you would be able to help me with my problem. Well here's the sitaution.

I have done a lot of recordings with my MP3 player off the FM and in each one have a buzz noise ever few seconds in the recording. I currently have been using Nero Wave Editor, SoundForge and Goldwave, but I can seem to get rid of of that constant noise. I have used noise reduction and several other filters on each and maybe I am not doing it right or something but I can't remove it.

In SoundForge I have used downloaded filters from digitalFAQ, also the noise gate, Click and Crakle and the noise reduction but nothing.

In Goldwave im currently trying the noise reduction but the audio does sound natural it sounds different maybe distored or damaged not sure.

If anyone could provide me with any hints or point me in the right direction on what to do I would appreciate it. If you need any additonal information let me know.

I have a included a sample of the buzz noise that I hear in the recordings.
Thank You

Noise in lecture

Sample of Buzzing Noise

How does this sound to you? Noise Reduced Clip.

This was decoded to wave file, loaded into Adobe Audition, noise profile captured where buzz occurs within silence, noise reduction using captured noise profile applied to whole clip, re-encoded to 256kbps mp3. The sound is altered, which is only to be expected, but the annoying buzz is gone.

First of let me say thank you for taking the time to do that for me I really appreicate it. So you are saying that I first make the audio a wave use buzz as a noise profile and then recode it is that what you are saying?

Any information you can provide would be greatly appreciated.

I don't know of any way of doing this without first decoding it to wave file. If you want a 'free' solution, the wave editor in Exact Audio Copy also does the job pretty well. You select the area to create the profile, then select the whole wave and reduce the 'noise'. In this case I found it needed -18dB to eliminate the buzz at normal listening levels, but you can play with that to see where it needs to be for your requirements. After that, do a 'save as' so that you retain the original and then you can encode using whatever codec you wish. If this is all spoken voice, 256kbps is something of an overkill, I would try something much lower, 96kbps, or even lower, should suffice.

As john33 hasn't replied yet, I will [edit: d'oh, trust me to get sidetracked for 10 minutes while john33 replies! Turns out we said much the same thing]. Yes, with the algorithms called "Noise Reduction" in almost every WAV editor (even the one provided in Exact Audio Copy) operate by taking a "noise profile" sample with just the noise you wish to remove. The longer the "noise profile" sample (usually a second or two), the better defined are the statistics used to determine just how much noise to remove and how much to retain at each frequency bin, as it's assumed to be the desired signal when it's above a certain level but noise if it's below that level.

Different algorithms perform differently especially with regard to damage to the original sound or addition of sparkling or sci-fi computer burbling in place of the noise. Most allow you to adjust the strength of the reduction to achieve the best compromise.

If you want to try one for free, EAC and Audacity are worth a look, and I think there are one or two command-line executables for this task.

Providing the buzz or noise doesn't change very much in frequency or amplitude throughout the recording compared to how it was in the noise profile, the noise reduction should allow you to make a lot of noises a lot less annoying without massive damage to the sound you're trying to keep.