Hi, I am just wondering if anyone knows of a program that can take in a "badly" recorded sound to output a "good" sound. For example, imagine you recorded a song over the telephone, the result would sound terrible, would there be some way of improving the sound? The specific thing I am refering to is music that has been recorded from a fair distance away, it was sampled at a high quality, but it sounds like it was recorded at 8khz. The bass is still fairly good, but the mid range is average sounding, and the high range is terrible. If there isn''t a program to do this, does anyone have any other resources to help "reverse" the bad-sound? I know for image processing if you have a blurry image, and you can determing the filter that was used to blur the image, you can pritty much un-blur the image. Is there a similar thing for audio? Any advice on how to construct the "unblur" (or "unbadsounding" in this case) filter? Thanks.

A crappy sound can be passed off when mixed in with good sounds.

You can try and make a crappy sound more pleasing to the ear by playing with it''s EQ.

Ultimately, a crappy sound will always be a crappy sound.

It sounds like this music was recorded without being monitored. The sound that your body perceives and the actual sound captured by a microphone are two very different things; never assume that because you can hear something clearly that a microphone can also hear it with the same clarity. To get a good recording you need to monitor what is being recorded by the mic via the headphone output of your recording device.

Better luck next time

Thanks for the reply,
Yeah playing with the EQ has made it sound a bit better, is there anything else that can be done?

What actually hapens when someone "digitally remasters" something? Do they do anything other than play with the EQ?

Its not really easy to re-record the sounds. (get an opertunity to about once every month-or-so)

sounds like you''ve got a high noise-to-signal ratio

one of the free v.s of cooledit (2000?) archived on the net ought to help.. might get good results just using a combo of eqing, noise reduction.. takes practice, ie. finding freqs to eq that will help the noise redux on your sample..

one basic ''trick'' you might not know is adding inverted signals.. oc if you invert something and add it, you get zero.. consequently, if you low-pass a signal, invert it and add it to the dry, you''re left with the high end.. so you might ie. isolate the sounds you don''t want, invert, and see if that gets you closer..

one synthesis technique is waveshaping to add even harmonics.. ie. 1 octave above, 2 octaves, et c.. this can be a way of boosting the highs w/o sounding too unnatural.. only used it on pitched/musical stuff so far tho.

but yeah.. good source is the best

Unfortunately cool edit is discontinued now, but I will try and find some other free software. Thanks for the advice, hopefully that way I should be able to remove some of the "noise" while preserving and amplifying the high-end.

what is "waveshaping"? iirc there are many different types of "waveshaping" ? could you be more specific? (or mention the app/plugin you are thinking of?)

http://www.tucows.com/preview/193466.html

didn''t follow that but ought to find ce 96 archived at dancetech.com

waveshaping would be signal processing, afaik the vernacular use implying amplitude operations, ie. you can turn a tri wave into a sine wave with waveshaping (amplitude transformation) but not into a saw wave

more a loose thought.. if it was really good for ''refidelitizing,'' there''d prolly be someone capitalising on it.. but if you want to give it a go, try www.synthedit.com - free

not thinking mathematically (ie. i dunno if it works in theory.. only in practice.. using the formula 5*cos(x) [where x=waveform -5 to +5..] will boost even harmonics in the ws2 module..

well.. i reckon every human who uses electronics should use synthedit.. but not sure if really worth your time..