Skip to main content ↑ ↓

Notice

Please note that most of the software linked on this forum is likely to be safe to use. If you are unsure, feel free to ask in the relevant topics, or send a private message to an administrator or moderator. To help curb the problems of false positives, or in the event that you do find actual malware, you can contribute through the article linked here.
Topic: Best programs or methods to clean up cassette tapes? (Read 19343 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

Best programs or methods to clean up cassette tapes?

Hey guys. I am doing a project for a friend, and to this forum's recommendations I picked up a TEAC  v-1050 to get the best sound from my tapes for a reasonable price. I now have them on my computer in WAV formats and am trying to get them to sound as good as possible. I mess with the EQs a bit and stuff like that but nothing sounds great. Some of these cassettes sound so good I don't need to do any adjusting, some are pretty poor and 13 years old and suffer from simply a flat sound, a bit of garbling (usually the sound you get when you physically wrinkle the cassette tape but this didn't happen) and hiss.

I was wondering what programs you guys generally recommend for this stuff. I am using Windows XP and tried Cool Edit but I found the de-hissing removes too much sound I'd otherwise rather have in. I'd appreciate any help big time as to what you guys like to use for this stuff. Thanks!

Best programs or methods to clean up cassette tapes?

Reply #1 –
Quote from: DaddyLongLegs on
Hey guys. I am doing a project for a friend, and to this forum's recommendations I picked up a TEAC  v-1050 to get the best sound from my tapes for a reasonable price. I now have them on my computer in WAV formats and am trying to get them to sound as good as possible. I mess with the EQs a bit and stuff like that but nothing sounds great. Some of these cassettes sound so good I don't need to do any adjusting, some are pretty poor and 13 years old and suffer from simply a flat sound, a bit of garbling (usually the sound you get when you physically wrinkle the cassette tape but this didn't happen) and hiss.

I was wondering what programs you guys generally recommend for this stuff. I am using Windows XP and tried Cool Edit but I found the de-hissing removes too much sound I'd otherwise rather have in. I'd appreciate any help big time as to what you guys like to use for this stuff. Thanks!


Try CoolEdit agin but not 'hiss reduction'. Find a non-program noise only section and grab a short section of 'noise reduction profile' and then try 'noise reduction'. That should do a better job.

G²

Best programs or methods to clean up cassette tapes?

Reply #2 –
Audacity also has a similar feature where it can analyse a section of 'silence' on the tape and construct its own custom filtering. You can then adjust the depth of the custom filtering it applies. I have no idea how it works or whether it varies in any way from the same process with CoollEdit, but it's free software so it's worth a try.

Cheers, Slipstreem. 

Best programs or methods to clean up cassette tapes?

Reply #3 –
After some intitial experimentation with CEP Noise Reduction, I typically use levels around 30% (or less) as I can live with a little tape hiss.

As for EQing, there are a few threads that have some useful settings.

My big problem with cassette transfers is the Dolby B encoding. Without recalibrating your tapedeck by ear for each tape, it produces noticeable breathing and pumping, especially on poorer condition tapes.

I looked into approximating (roughly) a decoder in software (a continuing saga) but it needs custom tools due to the variable band expander (varies with HF content in the encoded signal above 1kHz, I think...) that you need to decode it.

Leaving Dolby B off produces variably boosted mids and highs, leaving it on breathing and pumping and overly attenuated highs.

After my initial experiences with  tape transfers, I am firmly in the less is more camp in terms of restoration.

Best programs or methods to clean up cassette tapes?

Reply #4 –
Quote from: audioapprentice on
After some intitial experimentation with CEP Noise Reduction, I typically use levels around 30% (or less) as I can live with a little tape hiss.

As for EQing, there are a few threads that have some useful settings.

My big problem with cassette transfers is the Dolby B encoding. Without recalibrating your tapedeck by ear for each tape, it produces noticeable breathing and pumping, especially on poorer condition tapes.

I looked into approximating (roughly) a decoder in software (a continuing saga) but it needs custom tools due to the variable band expander (varies with HF content in the encoded signal above 1kHz, I think...) that you need to decode it.

Leaving Dolby B off produces variably boosted mids and highs, leaving it on breathing and pumping and overly attenuated highs.

After my initial experiences with  tape transfers, I am firmly in the less is more camp in terms of restoration.


Are you playing the Dolby encoded tape on a machine with no decoder? When operating properly, Dolby makes the machine 'a little more than it is' - meaning if it had a little high end roll off, it becomes a bgger rolloff. Similarly a small boost becomes a bigger boost but if it was flat, it stays flat.

I haven't had my Nakamichi 580-M out of the box in years but I have no desire to fool with it. I SHOULD dust off the ope reel deck. I re-capped it and the Dolby processor a few years ago but haven't transferred any of the old tapes yet. I'll probably have to take the tapes to work and bake them in the food dehydrator before I can run them. Hey, they're _only_ from '72 - '79.

G²

Best programs or methods to clean up cassette tapes?

Reply #6 –
Quote from: DaddyLongLegs on
...some are pretty poor and 13 years old and suffer from simply a flat sound, a bit of garbling...


I'm assuming that you made the necessary azimuth adjustments?

Best programs or methods to clean up cassette tapes?

Reply #7 –
I provided some detailed instructions on CoolEdit’s NR use in at least two thread on this forum.. Results greatly depend on how you go about it.

Also look at the recordings in Spectral View. You may find you can reduce hiss on some with a low pass filter because there is no audio at higher frequencies. I think I would tend to do the NR first, but I don’t recall experimenting to see if it makes any difference whether you NR and then low pass, or vice versa.

Best programs or methods to clean up cassette tapes?

Reply #8 –
Quote
I am using Windows XP and tried Cool Edit but I found the de-hissing removes too much sound I'd otherwise rather have in.


Sounds to me like you've got the dehisser's settings too aggressive - assuming you managed to get a decent sample of the hissin the first place that is. Have you tried reducing the "Reduce Hiss By" amount or raising the "Noise Floor Adjust". The other settings - FFT Size, Precison Factor etc all have some effect too. That's often the problem with CoolEdit. It has so many variables to set that it can take a lot of messing around to get it right. However, I've always found the Dehisser achieves better results than the Noise Reduction filter.

Best programs or methods to clean up cassette tapes?

Reply #9 –
There is no one single program for excellant audio recovery,  there is no 'one technique fits all'.  As you have already noticed each recording has different particularities.  Other persons might not be as observant as you are and would not be asking these questions at all,  so I must believe that you would rather do more than less.

I generally work with 2 different programs,  Pristine 2000 or 2005,  and Goldwave.  Audacity can be useful.  Good software filters providing samples of clean noise are available on the net and are an important part of this process.

Use headphones,  and good ones.  As far as dolby noise reduction goes you are mostly speaking of an allignment issue.  I utilize the proper type of dolby (B,  C,  etc) and I turn up the treble on my amplifier,  put on my headphones and carefully adjust the allignment of the playback head on the cassette deck for each cassette tape for the best high-end frequency response.  You might want to turn down the bass on your amp while you are doing this.

Next,  don't forget to flaten out your equalizer settings before you continue.  Listen from your computer as you record, at least as you begin to record, with the headphones.  Listen before you begin for buzz,  hum,  in case you have a defective connection to your computer.  Any noise that you inadvertently might add to your recording as you transfer it to your hard drive will make your work much more difficult. I record in wave format at CD quality.

Correct the 'dc offset' on your recordings before you proceed with any other steps of your work,  and prepare to be very patient.  You will have to learn which noise is acceptable to remove,  and I suggest that you don't approach noise as a single object.  Noise often needs to be removed in multiple steps. There is so much to learn.  Listen back to your work at each stage and be prepared to reverse the project.

Your software is not a total solution.  You are very much a part of this process if you are intent on converting some good recordings,  but I have done very much basicly with my own knowledge, experience  and the software listed above.

Best programs or methods to clean up cassette tapes?

Reply #10 –
Thank you much for the replies. I am messing with a lot of EQ settings and such, just adjusting them as much as possible in Acid. I am definitely getting good results. I was not *just* trying to remove hiss, but also a lot of the stuff which goes with an old cassette such as it sounding more muffled.

Quote from: odious_m on
Quote from: DaddyLongLegs on
...some are pretty poor and 13 years old and suffer from simply a flat sound, a bit of garbling...


I'm assuming that you made the necessary azimuth adjustments?


Well, this is a factory refurbished TEAC so I'm assuming that would have been set already properly, no?
I bought this model on a recommendation here, but I am a bit paranoid because the recommendation came with the poster's opinion that dual capstan was not necessary. It's a 3 head system, but now I'm of course wondering if the dual capstan would have helped with some of the garbling.

Here's an example:

http://www.onyxdomain.com/topsecret.mp3
(ignore the second song in the clip, it's from a DAT so it sounds great)

The first clip is from a song from a cassette and you can hear a lot of garbling. It's not horrible, but it's there. I was under the impression this is usually from the tape being literally wrinkled or crumpled (like when it gets jammed and becomes exposed) but that has not happened to this tape. Would adjusting azimuth help this? Would a dual capstan system have helped this?

I really appreciate the replies so far!

Best programs or methods to clean up cassette tapes?

Reply #11 –
If the tape runs freely in the shell, the pressure pad is intact, and the tape doesn't appear to be physically 'eaten' in any way then a dual capstan deck would be unlikely to remove the gargling effect. It sounds like the effects of poor storage and/or a deteriorating shell/pressure pad to me personally. I'd be inclined to swap the spools out into a known-good shell and fast forward and rewind it then try it again. It may just be "the way it is", in which case you'll have to live with it.

If all tapes sound gargled then I'd be inclined to take a look at the capstan. If it's bent or has a lump of tape oxide stuck to it then there's your problem. I've never successfully unbent a bent one, but tape oxide is easily removed with isopropyl alcohol on a cue-tip (or cotton-bud as we call them here in good old blighty). As the deck was sold as serviced this is unlikely to be the problem, but it pays to be clinical about these things if the idea is to archive from tape.

Regarding the azimuth, there is only one correct setting, but many manufacturers (including many leading high-end brands) failed dismally when it came to setting the azimuth correctly before their decks left the factory. I'd be inclined to slip the front of the loading door off and have a twiddle. There is usually a very sharply defined peak in HF performance when it matches the azimuth of the recording deck, so it's very easy to get it approximately right.

BEFORE adjusting the azimuth, I'd strongly recommend generating a 10kHz sinewave  (Audacity can do this for free) and recording it to a decent quality tape. This will give you a reference for adjusting the azimuth back to its present calibration point in the future.

To access the azimuth adjustment screw, open the loading door and grip it firmly by the sides then pull upwards. You'll see a horizontal row of three holes near the bottom of the exposed door chassis. You need the one on the right. It shouldn't need more than half a turn either way, so there's no need to go mad on it. A small watchmaker's cross-head screwdriver should do the job nicely. Put the tape in, hit play and adjust the screw on-the-fly for maximum HF output.

Something else that may help us is if you can provide a 30 second snippet of a song we might recognise recorded from CD with the deck so we can assess its overall performance and eliminate it from the equation. I doubt that you've been sold a duffer, but it wouldn't hurt to provide a sample. A fresh TDK-SA tape or similar would do nicely for this job.

Cheers, Slipstreem. 

Best programs or methods to clean up cassette tapes?

Reply #12 –
Quote from: Slipstreem on
BEFORE adjusting the azimuth, I'd strongly recommend generating a 10kHz sinewave  (Audacity can do this for free) and recording it to a decent quality tape. This will give you a reference for adjusting the azimuth back to its present calibration point in the future.

Very good advice!

Quote from: Slipstreem on
A small watchmaker's cross-head screwdriver should do the job nicely.

I’d additionally recommend using a plastic screwdriver, not one of these (often magnetized!) cheap metal ones.

(And I like the advice given here very much: Correct the problem(s) at the source, don’t try to »electronically enhance« bad input. As we programmers say: »Garbage in—garbage out«. )

Best programs or methods to clean up cassette tapes?

Reply #13 –
I don't know if it's better in newer versions, but in Cool Edit Pro 1.2a, the noise reduction is more like an artefact generation tool!

Sonic Foundry (or whatever it's called now - Sony probably!) NR-2.0 has a nice denoiser, especially mode 2.

Tape Restore Live (linked above) is great -
1. do the transfer without Dolby
2. use tape restore live to correct the azimuth (you must still do the best you can on the deck itself - this stage is fine tuning, not a substitute) ...
3. use tape restore live to correct the bias (if it's a problem - I think this is tricky to get right though)
4. run SF NR2 to de hiss gently, then
5. apply Dolby using Tape Restore Live.

I wish Tape Restore Live did Dolby C as well. I wish it was DirectX - the above workflow makes heavy use of Winamp's diskwriter!

Cheers,
David.

Best programs or methods to clean up cassette tapes?

Reply #14 –
Quote from: Slipstreem on
If the tape runs freely in the shell, the pressure pad is intact, and the tape doesn't appear to be physically 'eaten' in any way then a dual capstan deck would be unlikely to remove the gargling effect. It sounds like the effects of poor storage and/or a deteriorating shell/pressure pad to me personally. I'd be inclined to swap the spools out into a known-good shell and fast forward and rewind it then try it again. It may just be "the way it is", in which case you'll have to live with it.


The MP3 clip I provided is from a very clean shell and pressure pad. I kept buying TDK SA's on eBay but wasted about 30 dollars because it was a crapshoot with which ones had screws. I ended up just buying a bunch of generic ones. They're brand new, and even though they are generic I can't imagine that would matter, as long as the pressure pad is brand new too. Also not all tapes sound garbled, just the ones I know are older. Some tapes from 1998 sound awesome. If there is "garbling" it's usually for like a millisecond (sounds like the tape cuts out for a billionth of a second). I'm assuming that's the recording and there's no way to fix that.

Is dual capstan better with no pressure pad better than a single capstan with a brand new pressure pad?

Quote from: Slipstreem on
azimuth


I will definitely give this a shot, and let you know how it goes. In fact I think you're the guy who recommended this deck to me! Hoping for the best!

Best programs or methods to clean up cassette tapes?

Reply #15 –
Quote from: DaddyLongLegs on
Is dual capstan better with no pressure pad better than a single capstan with a brand new pressure pad?
As long as there's no visible sign of wear on the playback head, ie, no worn groove where the tape runs, not necessarily. A good pressure pad will ensure that the cassette tape operates as originally designed and intended. Having heard inconsistent results from a few dual capstan Nakamichi's when playing back less than physically pristine tapes, I'd trust a fresh pressure pad against a healthy head more personally. It may not be the theoretical ideal, but it's often preferable in practice.

Cheers, Slipstreem. 

Best programs or methods to clean up cassette tapes?

Reply #16 –
To use dolby decoding or not is always a question,  but after working with hundreds of tapes,  I mostly have opted to use the correct version with careful head alignment.  In almost every aspect of this technical and in effect,  rather artistic area of work,  I would say never say always and never say never.

If your cassette tape is of an out of print vinyl lp I might mention that Pristine contains an excellant pop and scratch remover.  I suggest applying a lot of patient experimentation with those settings and again,  use the headphones. Remove some of the scratches and go back to other areas of noise before you continue to remove more pops and scratches.  The reason headphones continue to be so important is that it is extremely easy to over process different problems in a recording.  They may sound great in your room environment... (Rooms are noise maskers in their own right). Put on the headphones and you may suddenly discover that Etta James is sounding a bit psychedelic  in quality.  Better to catch your first stage of over processing and back up to the last quality file before you waste hours of careful work.

Goldwave can be used to sample specific noise to it's clipboard.  Sometimes it is more effective than Pristine.

I generally use my cassette deck only for transfer these days and the heads do get adjusted for each cassette by my own hand,  but the suggestion of having a backup head alignment tape mentioned here is of course a good one.

Best programs or methods to clean up cassette tapes?

Reply #17 –
Unless all these tapes were recorded on the same deck, and it was very stable mechanically, best results come from adjusting the azimuth separately for each cassette. Some picky people will do it a number of times over the length of a single cassette.

While it is easy to generate artifacts with CoolEdit’s NR, there won’t be any if you do it properly. That’s why I made the comment above about looking up the instructions for proper use.

Best programs or methods to clean up cassette tapes?

Reply #18 –
Quote from: Arnold B. Krueger on
Quote from: audioapprentice on
Leaving Dolby B off produces variably boosted mids and highs, leaving it on breathing and pumping and overly attenuated highs.


You can download software that does a pretty fair job of removing Dolby B encoding from:

http://www.hansvanzutphen.com/tape_restore_live/

Wow, havent seen this one before. I bet the trading communities would love it. I'm using the auto azimuth feature on on all analog sourced music and I feel it restores HF/Impulse responce in a nice way.

Best programs or methods to clean up cassette tapes?

Reply #19 –
Quote from: markanini on
Quote from: Arnold B. Krueger on
Quote from: audioapprentice on
Leaving Dolby B off produces variably boosted mids and highs, leaving it on breathing and pumping and overly attenuated highs.


You can download software that does a pretty fair job of removing Dolby B encoding from:

http://www.hansvanzutphen.com/tape_restore_live/


Wow, havent seen this one before.


Check my post on this thread of 11/28/2008. Same link. ;-)

There is a relevant technology that has been applied to open reel tapes. It is possible to recover the origional bias signal that was put on the tape when it was recorded. Using this as a reference, it is possible to remove signfiicant amounts of wow, flutter, and other long and short term speed variations.

http://www.npr.org/templates/story/story.php?storyId=7489316

There is speculation that the following tool may have been used:

http://en.wikipedia.org/wiki/Max/MSP

Best programs or methods to clean up cassette tapes?

Reply #20 –
Well, I'd seen "Tape Restore Live!" before but hadn't taken its claims too seriously as I couldn't see how it could possibly work in a DSP on my PC. As I've just found out this evening, I was very wrong.

I have stacks of old reggae recordings dating back to the old Trojan Records days and before that are either still on commercial CDs or converted by myself to LAME MP3 at -V3. I know their pedigree, so to speak. I also trust my own hardware because it's a custom-built setup to serve as my personal home entertainment centre. It's tweaked for me and nobody else so I can easily differentiate between better or worse sounding (to me) generally speaking.

Playing these old tracks back in Winamp via this plugin was a bit of an eye-opener and an ear-opener, I can tell you!

Individual tracks have unique fingerprints in terms of the amount of azimuth correction needed to be applied on-the-fly, but they always follow exactly the same pattern...

1/ All recordings that have been mastered in mono ALWAYS return a null in the azimuth correction window with no correction deemed necessary.

2/ Tracks that require correction but aren't corrected ALWAYS sound 'wooly' in mono.

3/ Allow the correction to take place when deemed necessary and all tracks WILL sound fine in mono.

4/ ALL stereo tracks deemed to need no correction pass through the plugin transparently (enough for me) regardless of user settings for azimuth correction.

For completeness of the explanation, I'm using settings of a 40um limit and 5um/s.

So far I either gain something or lose nothing. Any difference is always big enough for me to detect with ease without needing to carry out blind listening tests with this particular source material. It is very real and my ears like what it does, so it's staying enabled by default for a while in Winamp to see how I get on with it and so I can keep fiddling with the many other settings.



So, what's going on here? Is it behaving like an FM noise reduction system in some way by reducing sideband imaging? Is it actually reducing the 'flutter' element of the 'wow and flutter' from the earlier generation tape recordings? It certainly sounds like it is.

Cheers, Slipstreem. 

Best programs or methods to clean up cassette tapes?

Reply #21 –
The azimuth fix is cross-correlating the two channels to determine the head misalignment, and then fixing it.

There are commercial processors that fix this too - e.g. there's one available from CEDAR.


As you say, it really works, and gives you mono compatibility that's not normally possible from recordings replayed with an incorrect or variable azimuth.

It's not doing anything to the wow and flutter.

Cheers,
David.

Best programs or methods to clean up cassette tapes?

Reply #22 –
Quote from: Arnold B. Krueger on
There is a relevant technology that has been applied to open reel tapes. It is possible to recover the origional bias signal that was put on the tape when it was recorded. Using this as a reference, it is possible to remove signfiicant amounts of wow, flutter, and other long and short term speed variations.
It definitely sounds like an interesting process (see also Plangent). Bias frequencies are usually found in the 120-200kHz range. If this technique turns out to be interesting and effective, it might be wise to digitize analog recordings using high sampling rates. Unfortunately there are not many ADC's that can handle >192kHz rates that would be required to record the bias signal. Even 384kHz could be on the "low" side. I also don't see DSD with its high levels of HF noise as a good choice, although it has been (still is?) promoted especially for archiving purposes.
Most Majors have their analog tapes digitized already and I don't think they used >96kHz rates and they probably don't want to redo the process "just" to capture the bias. According to Plangent, 96kHz is the minimum rate to be used for their process. For the moment you have to send them your tapes or soundfiles. I wish there were a commercial plugin or standalone version available. Perhaps some day ?

Best programs or methods to clean up cassette tapes?

Reply #23 –
Some of those are amazing...
http://www.plangentprocesses.com/examples.htm
...not sure how this works with 192k or 96k sampling - I'm guessing the bias frequency comes through as a sub-harmonic?

Cheers,
David.

Best programs or methods to clean up cassette tapes?

Reply #24 –
Those examples are indeed amazing, but have you looked into how far you can still get without sync-ing to the original bias tone?

Even ignoring the good 'Dolby B' NR playback simulator and the excellent azimuth correction, the FM Stereo hiss removal and AM mono tape hiss filters in Tape Restorer Live still strip out an enormous amount of rubbish from some of the tracks I feed it with (whilst leaving many untouched) and the end result is always either at least as clean or cleaner sounding to my ears with the filters enabled.

If it all works this well with no accurate clock to sync the DSP to the media, it soon makes you realise just how amazing the results can be if you have one.

Cheers, Slipstreem. 

PS Apologies if I'm teaching Grandpa how to suck eggs.