Transferring Live Choral Performances from Cassette to PC/CD

[Paul G]

I sing with an amateur mixed chorus (soprano-alto-tenor-bass). We are accompanied by piano and, sometimes, acoustic double-bass or electric bass guitar. Several of our gigs have been recorded – all microphone-in-the-audience jobs to cassette. I would appreciate guidance on how to transfer these recordings to CD via PC. These will not be straight archival transfers a la the Smithsonian Institute/Library of Congress “Save Our Sounds” project. Instead, I want to make these recordings more “listenable,” e.g., adjust sound levels, use EQ, add compression, reduce tape hiss (while losing as little musical information as possible); rather like cleaning up the Beatles’ Star Club tapes, although the sound quality for the most part is not as bad as that. I am not a music professional, but I do have a sensitive ear and am finding my way around audio editing software. (I recently did a competent transfer of my mono “Help!” LP (1982 Parlophone reissue).)

Details (lots!) follow.

My playback equipment is a Sony WA9ES cassette deck run through a Yamaha RX-777 receiver.

My PC is an IBM ThinkPad A30 with a 1.0 G Pentium III processor, 384 MB RAM, Windows XP, a pre-installed Crystal Audio WDM sound card, and a pre-installed Toshiba SD-R2002 DVD-ROM. When I perform sound recording and editing, I use a Logitech TrackMan Wheel mouse rather than the built-in TrackPoint. The receiver is connected to the PC’s line-in jack through a Radio Shack Gold patch cord, two RCA plugs to one 1/8-inch stereo plug.

My recording/editing software is Sound Forge 6.0. I will probably burn the CDs with Click ’n’ Burn Pro 2.0. (I prepared my “Help!” CD-R using Click ’n’ Burn and Click ’n’ Edit LE.)

The performances were recorded with a variety of equipment:

1. Some were recorded with a Sony Walkman Professional Stereo Cassette-Corder WM-D6C and a Sony ECM-959A electret condenser microphone using Maxell XL II and XL II-S cassettes and Dolby B noise reduction.

2. Some were recorded with a Sony Stereo Recording Walkman Radio Cassette-Corder WM-SR1 with the accompanying stereo plug-in microphone using Maxell UR cassettes and no noise reduction.

3. Some were recorded with an Aiwa Walkman-type stereo cassette recorder with the accompanying stereo plug-in bud-type microphone using Maxell UR cassettes and no NR.

4. Some were recorded with a Sony piano-key mono tape recorded with built-in microphone using Maxell UR cassettes and no NR.

The performances were given in a variety of rooms. Some of the recordings have a significant amount of room reverb; while I realize I cannot eliminate the reverb, I would like to give the chorus more “presence.”

The number of singers for each performance ranges from about 15 to 50.

The chorus as a whole was not miked through a house PA, but the soloists sometimes were. The accompanists were not miked.

If you’ve got this far, thanks for your patience! I look forward to hearing from you.

Paul

[Grant]

Cassettes are a bit easier to do, if you ask me. The hard part is making sure the heads are aligned to the tapes, and that the tape is running at the proper speed, and that they are properly decoded with Dolby, if they were encoded with Dolby.

Of course, you can do what you wish to your sound, but I tend to leave cassette recordings alone, aside from a slight bit of EQ. I do not clean up the hiss.

Bottom line for me, if the tape playback sounds right, all I need to do is cut the recording into individual tracks, and top and tail the tracks by fading and/or gating. Good thing you have SF 6.0. It is so fast and easy to do! But, since you have a variety of tracks to meld together, you have your work cut out for you.

I suggest that you first simply transfer all the songs to the hard drive. Then work on each individual track to make sure you get the sound the way you want it.

Once you finish that, you get to "master" the thing by getting all of the songs to "sit" together, or to make them all sound sonically even, with reason, of course.

Then, RMS normalize them to get them all to the same volume. I recommend the -16db preset in the Normalize tool in the Process menu. (It helps if you downloaded the seperate Batch Converter that comes free with SF!)

Once you are done, make your layout, and burn your disc with DAO, or disc-at-once.

[JoelDF]

I agree with Grant on the head alignment deal. It may mean having to snap off the cover of the cassette loading door (which usually can be simply snapped up and off). Then finding the little slot where the heads are and adjusting the little screw that holds the tapehead in it's place. Adjust it 'til you hear the sound change to the brightest it gets before dropping off and then go back to that bright part. If it's already at the brightest, then you'll hear it change by sounding more muffled - just adjust back.

Almost every tape deck is adjusted differently there. Even two commercially made tapes aren't always recorded on machines that are aligned the same - even though there is supposed to be a standard.

I've only transferred 1 tape to CD so far. I did do some NR to it because it was an old commercially made tape from '81 that did not use Dolby NR originally. It was a light amount of NR, but for my taste it needed it. It was a tape from a local band that was not of real high fidelity to begin with - and it spent 10 years of it's life sitting in the tape case on the back seat floorboard of my car.

Joel

[Grant]

If you adjust the heads this way, switch on the mono button on your amp first, to make sure you eliminate phase error.

[sgraham]

I agree with everything Grant says, except: since this is Choral music rather than pop, you may NOT want to normalize everything. Some pieces are meant to be quiet all the way through. But I'm sure you know all about that. Instead, I'd pick the loudest piece, normalize that so that the loudest peaks approach 0dB full scale, then use that as a reference to judge the volume requirements of the other tracks.

Compression is good if you plan to listen in the car, but it can really wreak havoc with this sort of music. I'd probably leave it alone, or make a separate version for car listening.

[Grant]

I missed the part about normalizing. I would normalize the whole think in unison but not the individual files.

[Paul G]

Grant, Joel, and Steve: thanks for your responses.

I have some follow-up.

1. I see the need for EQ on some of the performances to give them more "presence"; some were recorded reverberant rooms and sound a bit muddy. I realize I cannot actually eliminate the reverb. Can anyone suggest specific tweaks to make the sound a little more clear?

2. As for Steve's comment that "this is Choral music rather than pop" and what effect compression would have, perhaps I should describe the type of music we perform. Most of the songs we perform are pop songs, showtunes, and "standards," rather than "classical" or operatic pieces, so preserving the full dynamic range is not essential, in my view. I do not want a listener to be prompted to turn the volume up at the beginning of a piece and then, as the piece builds, have to turn the volume back down. Can anyone suggest compression settings that will relieve a listener of the need to continually adjust the volume during a particular piece but not completely destroy the dynamic range?

Thanks for your help.

Paul

[sgraham]

We're getting into dangerous waters here. I will offer just one piece of advice about compression/limiting: Use a fast attack and *very* slow release, for the least damaging sound - most like careful manual gain adjustment; or else use fast attack and fast release, for more squashed or "tightly controlled" (or "louder") sound. Anything in between is generally pretty useless, IMHO.

Do note that any compression or limiting is likely to make the reverb seem louder.