I taped a concert last night, I am trying to make the audio better, but like the topic says.....
Here is a clip, If anyone has an idea of how to make this sound better, let me know
The only thing I have done so far is to Maximize the volume.
http://us.share.geocities.com/xdvds/SampleSmall.mp3
Thank You
I have no idea what Im doing
-
Togglehead
- Posts: 98
- Joined: Mon Apr 10, 2006 1:37 pm
- Location: Jersey
-
piano nick
- Posts: 423
- Joined: Wed Jun 16, 2004 8:33 pm
Videofreak:
This recording hopefully demonstrates what a powerful computer the human mind is when connected to the human ear - I would make a guess that you thought the concert was pretty good, and didn't really notice the background noise until you played back your recording.
Judging by the music (not a criticism) I'd guess you are relatively young. I went for a hearing test on Thurdsay, and mentioned that fourty years ago, I never had trouble understanding people with heavy accents using the English language, and that I do now. What the audiologist said was very interesting: Although the signals from the ears to the brain may not have degenerated appreciably, the ability of the human mind to sort out the sounds coming from the ear does degenerate with time.
Consequently, when I went to a concert a few months ago, I found that when some audience members started whistling, cheering, and clapping before the song was finished, I was getting annoyed by them - I couldn't sort out the background noise from the concert.
Your recorder is not likely any smarter than I am, and it picked up everything that was "in the air".
You may play around with various filters, but keep in mind that the music frequencies you want are the same as the frequencies of the "background" noise. You filter out one, you filter out the other.
This type of recording is always crappy because you can't lug in a good reel to reel recorder and a bunch of expensive microphones - they would stop you at the door.
I have found a satisfactory solution - If I really like the group/soloist, I buy the CD's that they invariably sell after the performance. And I get it autographed by the artist. A CD is a lot cheaper than live most performances.
PN
This recording hopefully demonstrates what a powerful computer the human mind is when connected to the human ear - I would make a guess that you thought the concert was pretty good, and didn't really notice the background noise until you played back your recording.
Judging by the music (not a criticism) I'd guess you are relatively young. I went for a hearing test on Thurdsay, and mentioned that fourty years ago, I never had trouble understanding people with heavy accents using the English language, and that I do now. What the audiologist said was very interesting: Although the signals from the ears to the brain may not have degenerated appreciably, the ability of the human mind to sort out the sounds coming from the ear does degenerate with time.
Consequently, when I went to a concert a few months ago, I found that when some audience members started whistling, cheering, and clapping before the song was finished, I was getting annoyed by them - I couldn't sort out the background noise from the concert.
Your recorder is not likely any smarter than I am, and it picked up everything that was "in the air".
You may play around with various filters, but keep in mind that the music frequencies you want are the same as the frequencies of the "background" noise. You filter out one, you filter out the other.
This type of recording is always crappy because you can't lug in a good reel to reel recorder and a bunch of expensive microphones - they would stop you at the door.
I have found a satisfactory solution - If I really like the group/soloist, I buy the CD's that they invariably sell after the performance. And I get it autographed by the artist. A CD is a lot cheaper than live most performances.
PN
-
videofreak
- Posts: 2
- Joined: Fri May 12, 2006 12:23 am
-
Togglehead
- Posts: 98
- Joined: Mon Apr 10, 2006 1:37 pm
- Location: Jersey
-
Blandine Catastrophe
- Posts: 253
- Joined: Sun Jul 24, 2005 3:25 pm
Hard to extrapolate without to have heard (mp3 downloads are forbidden here), so I try to guess what problems can occur from the informations you wrote...
To filter cheers, try bandstop filters around 2500 Hz, because the cheers timber is likely more nasal than the singer's voice. With the spectrogram, you should get an idea of what to filter out. Not granted it will work, but I don't see nothing else.
Whistles should be easier to reduce. See in the spectrogram at what frequency range they start and at what frequency range they stop, anf use a dynamic bandpass filter to reduce.
The best is to create about four copies of the song and to select parts you will mix with View/Specify/CTRL+W, and shape volume in order that the changes be progressive. That will allow you to start the filters a little bit before the part you need to filter and to stop it a little bit after, and with properly faded parts, to hide the changes.
In order to avoid phase problems, use the same filter on the whole thing, but on parts when you neeedn't to filter, select a bandstop setting that makes an unnoticeable effect all the long of the file. That will make the main track, the one you will save at the end.
On a part where you need to reduce a whistle, for example, filter on a copy the whistle with the bandstop, select a shorter part such in a manner that you can select exactly the same part on the final track. Shape volume to fade the start and end of the selection, copy to the clipboard, and go to the final track. Select the same timing, do the invert of the previous fade (to remove the middle), and mix the clipboard at this place.
The other problem with concerts is the acoustic of the hall. Probably you'll need to reduce the bass range and add some treble, 6 and more 16 kHz, that will make something clearer.
For the claps, there's no way usually to filter them.
Depending on the parts, it can help to boost specific range to make an instrument to ressort. The graphic equalizer is like the band filters, there's a micro-delaying that can be an annoyance if not taken in account, so, for a progressive effect, it must be made on the same way with a main track who has been equalized, flat or not, but equalized anyways, and on copies you do the specifical equalizations for instruments and mix them on the same way that precedently.
Despite of all what you will attempt, the audio quality will remain poor if it is poor at the start. Sometimes the remedy can be more harmful that the disease.
To filter cheers, try bandstop filters around 2500 Hz, because the cheers timber is likely more nasal than the singer's voice. With the spectrogram, you should get an idea of what to filter out. Not granted it will work, but I don't see nothing else.
Whistles should be easier to reduce. See in the spectrogram at what frequency range they start and at what frequency range they stop, anf use a dynamic bandpass filter to reduce.
The best is to create about four copies of the song and to select parts you will mix with View/Specify/CTRL+W, and shape volume in order that the changes be progressive. That will allow you to start the filters a little bit before the part you need to filter and to stop it a little bit after, and with properly faded parts, to hide the changes.
In order to avoid phase problems, use the same filter on the whole thing, but on parts when you neeedn't to filter, select a bandstop setting that makes an unnoticeable effect all the long of the file. That will make the main track, the one you will save at the end.
On a part where you need to reduce a whistle, for example, filter on a copy the whistle with the bandstop, select a shorter part such in a manner that you can select exactly the same part on the final track. Shape volume to fade the start and end of the selection, copy to the clipboard, and go to the final track. Select the same timing, do the invert of the previous fade (to remove the middle), and mix the clipboard at this place.
The other problem with concerts is the acoustic of the hall. Probably you'll need to reduce the bass range and add some treble, 6 and more 16 kHz, that will make something clearer.
For the claps, there's no way usually to filter them.
Depending on the parts, it can help to boost specific range to make an instrument to ressort. The graphic equalizer is like the band filters, there's a micro-delaying that can be an annoyance if not taken in account, so, for a progressive effect, it must be made on the same way with a main track who has been equalized, flat or not, but equalized anyways, and on copies you do the specifical equalizations for instruments and mix them on the same way that precedently.
Despite of all what you will attempt, the audio quality will remain poor if it is poor at the start. Sometimes the remedy can be more harmful that the disease.
Gloup? :-°
-
piano nick
- Posts: 423
- Joined: Wed Jun 16, 2004 8:33 pm
-
john harding
- Posts: 4
- Joined: Wed May 17, 2006 6:48 pm
- Location: London
I'm a Goldwave novice but I did spend much of my life in sound broadcasting, one way and another, and I would say that this illustrates the fact that you can manipulate information, but you can't do much with information that you don't have. In this case, that includes high frequencies and multiple mics.
One of the first lessons you learn about recording things is that what you hear is not what you get, because what you hear benefits both from stereo discrimination and from the ability of the mind to ignore what it's not interested in.
One of the first lessons you learn about recording things is that what you hear is not what you get, because what you hear benefits both from stereo discrimination and from the ability of the mind to ignore what it's not interested in.