Recording - October 2012 - 50

B Y M I K E M E T L AY

Roland’s first software product, R-MIX, provides users with a set of interesting and intuitive tools for audio editing, repair, and remixing. It combines Roland’s established VariPhrase engine for formant-preserving pitch and time alteration with a new technology that Roland calls V-Remastering: a visual approach to sound editing, based on an animated display of frequency spectrum and pan over time. Finding your way around R-MIX is a standalone application that installs from CD-ROM with serial number authorization. It runs on Windows XP SP3 or later, or on Mac OS X 10.5.8 or later. (We didn’t review R-MIX Tab, which runs on the iPad.) R-MIX can import WAV files (plus AIFF on Mac) at 16-bit resolution and 44.1 kHz or 48 kHz sample rate. All files are exported as 16-bit/44.1 kHz WAV. R-MIX’s all-in-one-window layout is dominated by an animated color display called the Harmonic Placement screen. Buttons for basic and commonly-used functions (import, save, etc.) are above the Harmonic Placement screen, and controls relevant to the display (plus the VariPhrase pitch/speed controls) are on the sides. The bottom of the window has a time display, level meter, transport and function controls, and a dual-Track timeline/waveform display for the currently-processed file(s). R-MIX can play back two mono or stereo files at a time; these files are arranged on the two Tracks much as you’d do in a wave editing program, with point-by-point level automation. This lets you crossfade between files and create a playlist that can be exported as a single stereo WAV file, up to 60 minutes long. You can set up an A/B repeat region, place up to 20 markers, and take snapshots of the various control settings at the markers to automate what R-MIX does while creating and exporting your finished recording. The Tracks offer zoom, mute, tempo match, and click-and-drag removal of track beginnings/endings. There are few keyboard or mouse shortcuts, most functions have dedicated buttons. Harmonic Placement Now that we know how to import, arrange, mark, automate, and export our audio, all that’s left to discuss is what R-MIX does to these files! (So sue me, I saved the fun part for last.) The Harmonic Placement display shows the audio playback as animated blobs of light that change size, position, and color as time goes on. Color indicates how loud that part of the audio spectrum is, from cool purple/blue/green for quiet sounds to hot red/yellow/white for loud sounds. Sounds appear higher or lower onscreen depending on their frequency content, and left/right depending on pan position. The user places a boundary called a Frame around part of the audio—either a rectangle with adjustable size and shape or a stretchable oval/circle. Which one you choose depends on personal taste and on the audio you’re trying to isolate; I primarily used the rectangular box because I found it easier to tweak, but for certain well-localized sounds I found the oval provided a “tighter” fit. In the main screenshot, you can see my playlist of songs from Abbey Road by the Beatles; on the two Tracks I’ve placed “Her Majesty” right after “Mean Mr. Mustard” and restored the final
50
RECORDING October 2012

Roland R-MIX
chord of the song in the process, and I’ve placed the Frame around the very loud cymbal part that’s panned hard right. Once the Frame is placed, you have independent control of what’s inside the Frame and what’s outside of it. There are controls for cutting or boosting the level of the Inside vs. Outside audio, and setting their Pan positions; in addition, the Inside audio can have an Effect and/or Noise Cancellation applied to it. Once you understand how this simple system lets you isolate certain sounds in a mix and process them separately, the real fun begins! The basics and the not-so-basics The demo songs and Help files that come with R-MIX show you how to create a “music minus one” mix (e.g. for karaoke), alter a song to help you learn to play a part by ear, edit and clean up mixes, and create finished “medley” playlist files. The demos are designed to give you great results quickly (which is the point); realworld songs are likely to be a lot more dense and undifferentiated, making them harder—but rarely impossible—to work with. The “music minus one” aspect is pretty obvious; you find a particular instrument or vocal in the display, place the Frame around it, pull down the Inside Level slider, and most of the isolated signal is removed. Figure 1 shows how I did this for the acoustic guitar part in the song “Margaret Vs. Pauline” from the album Fox Confessor Brings The Flood by Neko Case. The song opens with the guitar playing nearly alone; that degree of exposure makes it easy to capture in the Frame. I used the A/B Repeat function to help me really nail the guitar as tightly as possible. Once that was done, I found I could boost or lower the guitar level in the full mix. I couldn’t make it vanish completely, and what I removed also altered the timbre of the remaining audio, but the effect was pretty dramatic and very easy to obtain. For learning parts (like guitar solos), you do the reverse: find and Frame the part you want to study, and then turn down the Outside level to remove the rest of the song. This is where the VariPhrase controls come in; you can shift the pitch and timing of the audio independently of one another, to place the part in a convenient key, tune it, or slow it down. The VariPhrase shifting sounds best at small intervals; while it won’t render obsolete the pitch/time tools in professional DAWs, let alone Auto-Tune or Melodyne, it works well for these

A new program gives users a new way to look at audio

Recording - October 2012

Table of Contents for the Digital Edition of Recording - October 2012