Audacity Forum

Hi to Audacity users: I'm relatively new to this piece of kit so I may be missing something, or there may be a plug-in you can suggest that will help.

I have a large collection of mostly spoken audio on cassette tape from the radio.

A fair bit of this is now kept by the BBC, so is not necessary to keep at home as well, but quite a lot is not available, so I have began to convert to mp3 and Audacity seems the obvious tool for the job.

I am having some problems, however. The 'stable' version had a noise removal tool which could not remove tape hiss and transformer hum without leaving voices with horrible breathless fluty tones. To some extent this could be reduced by sticking to mono, but it was all rather disappointing.

Then I tried the beta version with it's extra sliders on the noise remover, and with the frequency smoothing on high, most of the fluty tones are gone and the mono recording is now quite acceptable. I still can't get it quite right on stereo, but in most cases this is not important.

However, what I can't fix, is differences in tone from one side of the tape to the other. Typically, I will have one side of a tape that is nice and bright and crisp, and the other is dull and muted by comparison.

If I take out the tape housing from the player, I can manipulate the tape position relative to the heads and make slight adjustments in crispness/brightness that way, but with some tapes one side is dull no matter what I do.
[Possibly I could see what happens if I reverse the phono leads each time I change sides....]

I have tried the equaliser tools, but cannot seem to find any way of matching the brightness and voice tones (with the same person speaking), of an 'A' side with those of a 'B' side.

It seems all I really need is an overall tone control like those traditionally used on audio equipment.
Failing that, some way of analysing the overall tone and brightness of one side, and applying it to the other.

Is there any way to do this? Rather irritatingly most of the small portable devices I would listen to this material on don't have tone adjustments themselves, so I have somehow to get it right at the transcription stage.

I see there is a 'bass booster' filter: why no 'treble control'?

Anyone have any ideas about how to get both sides of a tape transcription sounding as if they come from the same speaker, on the same day, in the same studio?

Cheers,

Spamlet

You state you have tried the equalizer. I assume you find that it has an effect, but you are just having trouble matching the sides to sound about the same and that you are doing this by trial and error?

How about applying a low pass filter to both sides of the tape? That should make each side approximate the other in the higher frequencies. You'd have to experiment on the low pass setting.

Then apply the same equalizer settings to both sides, boosting treble on both sides?

In the equalizer, you can select any of a bunch of curves from the "select curves" dropdown. It looks like most or all of them increase bass and reduce treble-----but there is an "invert" button on the right side that would instead boost treble and reduce bass. I've never used the invert, but I'm guessing it is self-explanatory and would boost both sides of the cassette equally, after they have been subjected to the low pass filter.

Then possibly apply bass boost if needed to compensate for the reduced bass resulting from the inverted curve.

This might give you a less than ideal "brightness" on both sides, but both sides might be equally less than ideal. That might be preferable to having them differ considerably. It would obviously take some experimentation.

You can do a "plot spectrum" from the analyze menu, which will give you a frequency analysis graph, but I'm not sure how that graph could then be applied to your case or to equalizer settings.

Thanks for the tips Ignatz.

I've done quite a bit of experimenting with the equaliser settings, but none seem to be able to change this underlying muffledness of one side relative to the other, and I don't want to spoil the good side with any unnecessary processing, as it all affects the voice: changing it from the right sound for the person one knows. Now if I could somehow use this tool to extract a profile from one track and apply it to the other: that would be what I need. Perhaps there is actually a setting to do this and I just haven't found it yet.

Otherwise a simple tone control might do the job.

The analyser tool only works for very small sections of audio, and there is no indication of what one can do with the graph once one has it. Now if there was one that could analyse the salient characteristics of a whole track and then apply them to another, in the way that the 'noise' sampler does, we might be on to something!

Thanks for your interest,

S

Are you up to tweaking the head azimuth for each side of each cassette? There some pointers here: http://wiki.audacityteam.org/wiki/Recor ... m_Cassette

I've done this, and listening to the playback while adjusting the azimuth is easily done. [Some may gasp in horror and insist one needs sophisticated technical gear to do this right] All you need is a small Philips head (star) screwdriver.

You may find this is much easier than trying to compensate after the fact. Boosting the high frequencies for the side B recording will also boost the hiss.

-- Bill

Yes, that's more or less what I was doing by taking the cassette cage out of the recorder and moving the cassette back and forth rather than the pick up head (This is also handy when dealing with the cassettes that are so badly spooled that the tape often loops round the capstan and has to be grabbed back quick!).

Not sure I can get at the head adjustment itself without a complete strip down, but thanks for the reference.

Cheers,
S
(Impressively quick responses I'm getting here - but I'll have to go to bed soon)

Spamlet wrote: Not sure I can get at the head adjustment itself without a complete strip down, but thanks for the reference.

If you have the usual hi-fi front loading deck (as opposed to a slot-loading car-type deck) you should be able to get at the azimuth adjustment by removing the "decorative" front plate of the cassette holder. They usually slide off, and it may take a bit of tugging as they are designed to snap into place.

-- Bill

Yes, that is what I said I already have done to facilitate cleaning and untangling.
I hadn't noticed a head adjuster screw on this one - though I have done it on older models.
(I've even made my own drive belts before now - and made my own winding devices: it being pretty exhausting untangling a c120 the cat has got at; cutting the spirals into usable sections; sticking them together in order and the winding them onto fresh spools!)

(Incidentally, I was quite pleased to find, on dealing with tape again, that the 'special' white splicing tape I used to have to buy in little rolls appears to be the same bar colour as what is now called parcel tape. Splicing is now easy and costs nothing! though my finges are not as steady as they used to be!)

I'll have a closer look fot head adjustment.

But still think that for non technical users, Audacity could do with a straight forward tone adjusment.

Cheers,
(It's 4 in the morning here now, so must put this away!)

PS: SUCCESS with the head adjustment - even after I'd gone to bed! My tape recorder was better designed than I had thought, and a small hole that is only accessable with the tape cage removed, does indeed line up with the head adjuster screw when the play button is depressed. Thus one can adjust each tape individually while it is playing and get back some of the treble that was being lost (Looks like I've got yet another marathon effort of retranscribing to do: I practically can quote the tapes word for word after all these experiments!)

So a huge bunch of thanks to those who thought to point out the importance of the head adjustment. In fact it is so important it seems almost incredible that something recorded on one piece of equipment could ever be played successfully on another, without a head adjustment knob being provided right next to the play button!

Anyhow, this has enabled me to sharpen up my last few transcriptions, but there are still, presumably, differences in the recordings themselves that could do with some tone tweaking, so I'm pleased to see the suggestions are still coming.

What a very helpful group this is, so thanks very much to all!

Cheers,

S

Spamlet wrote: But still think that for non tecnical users, Audacity could do with a straight forward tone adjusment.

Have a look at this thread for a tone control plug-in

http://forum.audacityteam.org/viewtopic ... 20&p=67417

-- Bill

thanks Bill,

That's an interesting thread: though some of the discussion is still over my head.

Unfortunately, this laptop I'm using only has a small amount of ram, and the discussion goe into needing gigabytes of it.

Did download the add on and give it a try, but even selecting a few decibels treble increase it sent the computer into neverland, and Audacity had to be disabled.

So, it's still not the simple tone control I was looking for, though on a newer laptop or pc it may well do the job.

Still: I'm gradually improving my transcriptions thanks to all the help here.

Cheers,

S

Spamlet wrote:Unfortunately, this laptop I'm using only has a small amount of ram, and the discussion goe into needing gigabytes of it.

If you set "Normalize Output" to "Off" then the amount of ram required is considerably less.
It is the "Normalize" function that requires that the entire audio selection is temporarily loaded into ram. The tone adjustment itself requires very little ram.
If necessary the track may be amplified manually using Audacity's "Amplify" effect (in the "Effect" menu).