Sequencing Secrets: Fixing and Fattening Vocals

From Electronic Musician, November 1998

By Tom Pearson, audio director for the Minneapolis Metrodome.

John Mahar, production coordinator of the Minnesota Vikings, called me with a dilemma this past July. The Vikings wanted a new version of their theme song, Skol Vikings. John had all the tracks recorded, but the vocals sounded weak and were not in sync with the music. The song is played at every home game to get thousands of Vikings fans riled up after a big play. Big plays need a big song, and weak, out-of-time vocals just don't cut it.

"Don't worry, I can fix your problem fast with Cakewalk Pro Audio," I told John. "Just send me your tape."

The first thing I needed to do was fix the timing problems. I started by recording all the music tracks into Pro Audio. I then opened the Audio view and split the vocal track at the beginning of each phrase to create separate audio events. Once the vocal track was split into separate events, it was easy to drag the different phrases back and forth until they aligned with the beat of the music. When doing this, it's helpful to have the music tracks open in the Audio view, side-by-side with the vocal tracks. That way, it's easy to visually align the start of vocal phrases to the beat.

The next step was to fatten up the vocals. For this I used a process called "layering." This technique works great for any tracks that need thickening, especially vocals and rhythm guitars.

First, I copied and pasted the vocal track until I had six tracks of the same vocal part. I then used Pro Audio's Time/Pitch Stretch function to drop the pitch down 0.1 semitone on track 2 of the vocals, down 0.2 semitones on track 3, down 0.3 semitones on track 4, up 0.1 semitone on track 5, and up 0.2 semitones on track 6. By subtly changing the pitch of vocal tracks, you give the illusion of multiple voices singing the same part. Time/Pitch Stretch offers options for different audio types. For this application, I chose Vocal as the Source Material and set the Advanced Settings to High Accuracy and Formant Preserving Algorithm.

I now had a crowd singing, but it still sounded a little contrived. So to warm it up, I sent vocal tracks 2 through 5 to one of the aux buses in Pro Audio's Console View. Here I placed FX Stereo Chorus with the Thick Group Vocals preset and FX Stereo Reverb with the Arena preset on the tracks. I then sent the 100 percent wet submix to a separate output on my sound card.

With the wet and dry vocals combined, I now had the sound of a full stadium of raging fans singing "Skol Vikings." The song was ready. I sent it off to the Vikings' main office in Winter Park.

Then came the surprise: Winter Park called to say the song sounded "too new" and asked if I could try and make it sound a little more like the old version. (So much for state-of-the-art, slick production!) To solve the problem, I decided to add the old vocal track from the original version (circa 1970). But first I had some things to deal with.

When I recorded the original track into Pro Audio, there were two issues. The original recording was noisy and at a slightly different tempo. No problem. From within Pro Audio, I ran Sonic Foundry's DirectX Noise Reduction plug-in on the track. Then I used Pro Audio's Fit-to-Time command, and the old singers were singing right in time with the new ones.

There was still something missing, however. After listening to the original version a few times, I sensed that what was needed was more bass drum and tuba. I'm not much of a tuba player, so in came MIDI to the rescue. I was able to quickly add MIDI bass drum from a Korg NS5R and the tuba from a Yamaha MU90R. Now the song was complete. I sent it over to the Vikings and they loved it.

Now if the Vikings can perform as flawlessly out on the gridiron as Pro Audio did on this project, we'll see them at the next Super Bowl!

More Tips