I have a box of old cassette tapes (some are in great shape, some are multi-generation copies that degraded a lot) that I'm trying to convert into mp3s for permanent archiving, and I'm curious how others would approach such a task.
What I'm doing now is basically recording them as .wav and then apply noise removal, then a bit of EQ, and maybe a maximizer to tighten/clarify the overall sound. I don't apply limiters or compressors because I think that would mess up the original recording engineer's work. Does that sound about right?
Remastering old cassette tapes?
-
- KVRAF
- 1644 posts since 18 Mar, 2004 from Lincoln, CA
-
- KVRian
- 515 posts since 24 Sep, 2004 from Neverland
Yup, but Id use something like Spectrix or any other good enhacer/exciter for the highend !
And a maximizer is a good idea, since chances are youll be a few db away from 0, unless you recorded it as a teenager, and overdrived the volume like most of us did it back then
And a maximizer is a good idea, since chances are youll be a few db away from 0, unless you recorded it as a teenager, and overdrived the volume like most of us did it back then
-
- KVRAF
- 12977 posts since 29 Sep, 2003 from Ottawa, Canada
A Maximizer IS a limiter.
Sounds like a good procedure, though.
Greg
Sounds like a good procedure, though.
Greg
-
- KVRian
- 1219 posts since 12 Aug, 2002
Lunatique...my practice is to leave any noise reduction process till the last. The reason for this is because these algorithms rely on a spectral analysis procedure to identify noise elements in the frequency domain, and they will leave behind some artifacts, no matter how inaudible they may seem, based on the frequency content the algorithm has worked on.
It follows, then, that if you then proceed to re-balance that frequency spectrum, using post noise removal EQ procedures or dynamics processes, you run the risk of unmasking any noise reduction artifacts to the degree that they are now truly audible and, indeed, now intrusive.
I usually perform any tonal, and dynamics shaping first. Sometimes noise elements can be more tranparently reduced, or at least partly so, with shrewd EQing, using good tools. Often the tradeoff off of having a slightly noisier mix is preferrable to one which displays the less familiar artifacts left by noise reduction procedures.
For me, even the "really good" noise reduction algorithms, used wisely and carefully, still leave a certain "signature" behind on a music track that I am not particularily fond of, a certain depth and dimension is missing. In stereo files, the stereo correlation suffers. For this reason I have to have a really compelling reason for using noise reduction processes on music files. Compared to the noise most folks are exposed to these days, surely a bit of tape hiss is not going to seem like such a big deal.
It follows, then, that if you then proceed to re-balance that frequency spectrum, using post noise removal EQ procedures or dynamics processes, you run the risk of unmasking any noise reduction artifacts to the degree that they are now truly audible and, indeed, now intrusive.
I usually perform any tonal, and dynamics shaping first. Sometimes noise elements can be more tranparently reduced, or at least partly so, with shrewd EQing, using good tools. Often the tradeoff off of having a slightly noisier mix is preferrable to one which displays the less familiar artifacts left by noise reduction procedures.
For me, even the "really good" noise reduction algorithms, used wisely and carefully, still leave a certain "signature" behind on a music track that I am not particularily fond of, a certain depth and dimension is missing. In stereo files, the stereo correlation suffers. For this reason I have to have a really compelling reason for using noise reduction processes on music files. Compared to the noise most folks are exposed to these days, surely a bit of tape hiss is not going to seem like such a big deal.
To the mind that is still, the whole universe surrenders - Lao Tzu
-
- KVRian
- 901 posts since 1 Dec, 2003
I've been doing this for the past 2 years in my spare time. One thing I have had to tweak BEFORE going to .wav is the azimuth on the tape deck playback head. My cassettes were recorded on various machines over the years, and some of the decks weren't aligned, or they were just plain crappy decks with a loose transport.
Don't know if it's the same in your case.
I use a spare deck for this so I don't knock my good deck out of alignment.
Then it's Wavelab, Redunoise ,Harbal, and a touch of PSP Vintage warmer
Don't know if it's the same in your case.
I use a spare deck for this so I don't knock my good deck out of alignment.
Then it's Wavelab, Redunoise ,Harbal, and a touch of PSP Vintage warmer
-
- KVRAF
- 1577 posts since 20 May, 2002 from Cambridge, UK
hey I'm with you on the tape head alignment thing - I suggested it in another thread on the same subject only to be told it was a terrible idea (but I knew I was right 
)
)THIS IS MY MUSIC: https://spti.fi/rZyjX7i 
-
- KVRAF
- 2049 posts since 18 Sep, 2003 from Seattle USA
If someone has a better noise reduction tool than Redunoise (Voxengo) I'd like to know about it !
I'd even be glad to have a friendly shoot-out using a wave sample of any audio anyone chooses at high noon behind the OK Corral.
I'd even be glad to have a friendly shoot-out using a wave sample of any audio anyone chooses at high noon behind the OK Corral.
-
- KVRAF
- 1577 posts since 20 May, 2002 from Cambridge, UK
well soundforge or whatever it's called these days is the best I've used (too expensive though)... I rely on Acoustica for noise reduction now, though I don't know if it would be satisfactory for some people. Can provide an example if you're interested...
THIS IS MY MUSIC: https://spti.fi/rZyjX7i
-
- KVRAF
- 2548 posts since 7 Jul, 2003 from Huntington, WV
Why did they say it was a terrible idea?Jbravo wrote:hey I'm with you on the tape head alignment thing - I suggested it in another thread on the same subject only to be told it was a terrible idea (but I knew I was right
If you are using a deck dedicated for transcription purposes only, like Landphil does, I don't see a problem with it at all. What am I missing here?
take care,
McLilith
-
- KVRAF
- 1577 posts since 20 May, 2002 from Cambridge, UK
THIS IS MY MUSIC: https://spti.fi/rZyjX7i
-
- KVRian
- 1262 posts since 15 Feb, 2003 from Up the Pennine way
Do you know if the tapes were recorded with or without dolby.
What dolby does is it increases the top end on recording and cuts this on playback which dynamically reduces the tape hiss.
I'm asking cos I have done a load of tapes and found that if they were recorded with dolby I could clean them up better by having this turned off on playback.
If you want to adjust the heads then do it with one of the tapes you are going to "render" inserted. Turn the screw in 1/4 steps either way till you get the best output/sound. This is best done on headphones as you will easily hear the diference.
The only problem is that any recent tapes you have done
will have to go through the same process. Dart Pro is around $50 (i think) and is very good.
Audio Surgeon, $40 If you can find the demo it will give you 20 free goes.
http://www.compusa.com/products/product ... pfp=BROWSE
fake You cant beat people up then have them say "I love you"
-
- KVRAF
- 2548 posts since 7 Jul, 2003 from Huntington, WV
Regarding noise reduction, I see a lot of mentions for Cool-Edit noise reduction. Has anyone ever compared results from Cool-Edit noise reduction, with results from Audacity's noise reduction? I think Audacity gives practically the same results, but I'm not sure. I do know that Audacity is free, while Cool-Edit was bought by Adobe and now costs even more than it used to (especially since you can no longer get the basic 2-track version.) 
Also, I think using such noise reduction after altering dynamic range via compression, expanding, limiting, etc will give poor results. The noise reduction algorithms assume a fixed level for the noise. If you use dynamic range processors on the signal, the noise level will no longer be a constant volume, and the noise reduction can be thrown terribly off course. It would seem to me, that moderate noise reduction before dynamics processing would be the better approach.
I have some old cassette tapes made on cheap portable recorders which used AGC, and it's almost impossible to remove the noise on these tapes, because the level of the noise keeps changing. I always did hate the AGC used in cheap consumer recorders.
(AGC = Automatic Gan Control)
take care,
McLilith
Also, I think using such noise reduction after altering dynamic range via compression, expanding, limiting, etc will give poor results. The noise reduction algorithms assume a fixed level for the noise. If you use dynamic range processors on the signal, the noise level will no longer be a constant volume, and the noise reduction can be thrown terribly off course. It would seem to me, that moderate noise reduction before dynamics processing would be the better approach.
I have some old cassette tapes made on cheap portable recorders which used AGC, and it's almost impossible to remove the noise on these tapes, because the level of the noise keeps changing. I always did hate the AGC used in cheap consumer recorders.
(AGC = Automatic Gan Control)
take care,
McLilith
-
- KVRAF
- 2548 posts since 7 Jul, 2003 from Huntington, WV
For what it's worth, I adjust the head on my cassette deck before I transcribe an old tape. If I want to put the head into some sort of "standard" position, the best I have for calibration is commercial pre-recorded audio tapes. That's good enough for my needs.
I do find myself wanting a real calibration tape, though. Then, I could adjust the head, the tape speed, and more, to standard levels. It's getting hard to find calibration tapes. I can't find one for metal tape at all, only "normal" and "chrome" tapes.
Also, why on earth did they quit making metal audio cassette tape, instead of dropping normal or chrome instead? (They still make metal tape for 8mm video.) All my best cassette recordings were on metal tape.
take care,
McLilith
-
- KVRian
- 901 posts since 1 Dec, 2003
The response you got was valid if you own only one tape deck. After 20 tapes spanning 30 years, it's quite easy to lose the original alignment.
Which reminds me, has anyone successfully "decoded" a dbx type I encoded tape using plugs, such as filters, EQ's or compressors? I emailed dbx 4 years ago about developing a dbx plugin, but they said that the demand for such a plug wouldn't offset the cost of developing it.
I still have my type I rackmounts, but an "in the box" solution would be cool.