(ã 1998)
Articles are provided
Courtesy of:
Computer
Music Products
The
Making Of An Album
Page 5
- Recording The Lead Vocals (Digital Audio) -
by Michael Tyler
Computer Music Products
The day before Jeremy was to come to record his lead vocals, I decided to run a few tests to make sure everything would be ready to go. I intended to do both recording and playback of all digital audio using the AudioWerk 8 card. I connected the mic directly in the back of the card (using various plug adapters to convert the ¼" plug to an RCA-type for the AudioWerk). I then made a test recording of my voice in Cakewalk. The signal strength was too low to work without using a separate mic pre-amp. I then tried plugging the mic into a small desktop mixer (thinking this would boost the signal a bit), then routed its audio outputs into the AudioWerk. This was better, but I found I couldn't monitor the microphone's signal through headphones simultaneously with the MIDI instrumental playback using this setup.
Reluctantly, I plugged the mic into the SoundBlaster's mic input and made another test recording. I found I had plenty of signal and could monitor everything through headphones without any problem. I experimented with the record level inputs using the SoundBlaster's software mixer until I had an acceptable signal-to-noise ratio. I spent a few more minutes recording multiple digital audio tracks using the "pseudo" full-duplex mode of the SoundBlaster. The SoundBlaster AWE32 is not a true full-duplex card, meaning, it was not designed to record and play digital audio simultaneously. A software driver for the card forces the Blaster into a duplex mode. Although it records at 16-bit, what you hear in this mode, is an 8-bit rendition of your recording and playback. After recording, I disabled the "full-duplex" mode and was able then able to monitor the audio at a full 16-bit.
I still was not totally happy with the overall playback quality of sound, so I changed the settings in Cakewalk to use the AudioWerk card as the digital audio playback device. What a difference! The quality of sound was greatly improved and the signal-to-noise ratio was much better. I hesitate to say this, but compared to playback through the SoundBlaster, the audio sounded almost pristine! I was very surprised and relieved that I actually found that this combination worked. I must confess that there was one caveat, however. I discovered that there was a 30 millisecond timing difference between the SoundBlaster and the AudioWerk card. This is because every audio card has a slightly different internal "clock." Since I used the Blaster for recording, and the AudioWerk for playback, I had to compensate manually for this timing difference by shifting the audio recordings ahead in time by 30 milliseconds to keep synchronization with the MIDI tracks.
There was a setting in Cakewalk that I thought I could use to make the AudioWerk the "Timing Master" for both recording and playback. Unfortunately, it didn't make any difference. I finally resigned myself to the fact that I would simply have to manually shift the audio tracks throughout the entire project. This is not difficult (only takes a few mouse clicks), but it was extra work that I just as soon not have to do! But, again, I was glad that at least it was a workable solution. (note: E-magic later came out with a new driver for the AudioWerk that would have allowed me to use it as I originally intended, but it was too late for this project.)
Jeremy came over the following day, blissfully unaware of the problems I had worked out the day before, and we got down to business. I connected a set of headphones for both of us (using a "Y" adapter from the Mackie mixer’s headphone output) so we could monitor the vocal recordings simultaneously. I enabled a track for audio recording on each song for the lead vocal. After adjusting the input levels for Jeremy, we simply went from one song to the next, and recorded all the leads. After each recording, I moved the audio ahead by 30 ms so it was in sync with the MIDI tracks for playback. Everything went smoothly, and we got all the lead vocals recorded in just a couple of hours!
I created a new directory to resave these latest versions of the songs, leaving the "MIDI-only" versions alone. This was a safety measure so we could revert to the original versions if necessary (we never had to, by the way).
After Jeremy left, I processed each of his recorded vocals using the WaveLab software to remove DC offsets (slight signal mismatches between the two audio cards I was using). This made the vocals even cleaner sounding. I then resaved the songs in yet another directory, leaving the originals intact -just in case!
I left the vocals dry (no effects). We would add effects at some point after all the other vocals were recorded.
Next Page (pg. 6) Recording The Background Vocals (Digital Audio)
Back To Top of this article