The Sessions

The sessions ran for three days, with one day of rest after the first two days. We got sounds within two hours after the performers arrived at the studio, and then got right into it. We got seven songs in the can on the first day, on about 70 takes. The second day we got 10 songs done, in about 120 takes. We took a sorely needed day’s rest and then finished it off with four new songs and two repeats from the first day on our last day, in another 70 takes.

Naturally, these takes weren’t all complete! Nonetheless, the amount of singing was substantial for Kyoko. We tried to work at a fairly relaxed pace, recording for perhaps two hours (with lots of playbacks and discussions), break for an hour, record for two hours, break for an hour, and then record for another hour or so. By the end of the second day, Kyoko’s voice was beginning to sound a little tired, but after a day’s rest she was back in form.

The producer, Sachiko Nakada, exercised some very focused and careful control over the pace of the sessions, and as a result they went quite smoothly and with minimum stress. Language was a bit dicey, with long conversations over the talkback in Japanese, and the rest of us wondering what was supposed to be happening.

One extra plus here was the quality of the studio. Mark Conese very unobtrusively kept the lounge stocked with fresh food and coffee, the piano tuner popped in twice a day to touch things up, and the welcome mat was very much in evidence. The equipment ran flawlessly and Tom Bender did a great job as a tape op. So, we were really comfortable and we never had a busted take or noise problem – a really welcome change from most location work. The ST250’s were wondrously quiet and preamp noise was quite low as well. We were converting to 18-bit digital using a Mytec Technologies Reference A/D 2018 A/D converter and then truncating to 16-bit in parallel to two Tascam DA-30 DAT recorders. When I got back to Groton and started listening to the tapes, I thought I heard some low-level noise pumping. Uh-oh! I cranked up the level to have a really close listen, and what it turned out to be was the noise of clothing on skin as either Kyoko or Dalton moved! A new one for me! Really pretty quiet tapes!

One other oddity: the Genelecs yielded a strongly holistic image that I had trouble shutting off as I got tired. Usually, the strength of such an illusion would be regarded as a benefit and evidence of great speaker design. However, in this case, I had trouble hearing through it to “how the recording really sounded” so I had to switch to the big soffitted Augspurgers just to get my brain back in gear. Sometimes stereo is just too powerful!

Editing

Back in Groton, I loaded the probable takes for each song into a single song file in Digidesign’s Sound Designer II (SD II). Editing was then a comparatively simple matter of creating the appropriate Edit Decision List by creating regions from the desired takes and cut points, and then using the Edit Region Window to bring the edits into correct time and fiddling with crossfades until it sounded right. Because the noise floors were so benign, we had no trouble with continuity. Songs typically took only four or five edits. One song was a complete single take.

As I’ve discovered before, the tight little edit takes we tend to do to cover particularly difficult moments were hardly ever useful, because they always were too careful and polite. Usually it’s a matter for pulling from complete takes or large sections, going for the take that is “happening” best at any given moment. Occasionally I was called on to modify the musical phrasing or diction via use of creative crossfades and adding or deleting time. This is something that the Edit Region Window in SD II is really spectacularly good at.

As I finished editing each song, I dubbed it to DAT. The SDII files were dumped to digital data DAT (a DAT format used for backing up computer data that uses a special DAT drive) at the end of each day.

In post-production, I will load the edited songs back to Pro Tools for sequencing, little level fixes, and some minor EQ. This master file will then be printed to DAT with reverb. I’m using a fairly low-level, but long reverb time with considerable bass energy. I’m also adding a little spaciousness by boosting lows in the A-B component of the stereo signal. Ms. Nakada isn’t terribly concerned about mono compatibility, so I have considerable latitude in this regard.

Post a Comment