Two-Track Editors for the Windows PC

A plethora of new flash memory and minidisk recorders has hit the market, so there will be more stereo recordings coming in to your shop that never saw the inside of a DAW.
Author:
Updated:
Original:

Sony CD Architect 5

Sony's CD mastering program ("mastering" in the classic sense here, including writing and editing of subcode information) turns out to be a capable editing program in its own right. CD Architect 5 is bundled with Sound Forge 8 (or $120 as a stand-alone download). And while it's not a true object-oriented editor it behaves like one in many ways. You can load the file of a whole concert recording into the Time Line window, split it into tracks at logical points (typically the beginning of each song) and then the fun begins.

"Splitting" doesn't actually modify the audio file, but it marks start and end points for each segment and develops a playlist for the project. Each segment begins and ends with a fade region (10 ms default), which you can modify by dragging a corner of the segment or selecting a different fade curve. Creating a real fadeout or fade-in is as simple as dragging the "fade" corner of a segment out to the length you need. Those wanting more control of transitions from one segment to another can open a second "layer" and create complex crossfade shapes between segments on each layer by using volume envelopes.

CD Architect is quite a capable editor once you get the hang of it. You can cut out extraneous material simply by dragging the edges of the boundaries, adjust spaces between the tracks, or crossfade one track into another with no silence. This is a cumbersome way to rearrange speech, but it's fine for shortening a rambling introduction and putting songs in their final order. You can burn a standard audio CD directly from the program, and you can also render the assembled and edited material into a single file, incorporating volume adjustments or any plug-ins that you apply. Those with the need and knowledge can enter ISRC codes and export the PQ code list. CD Architect could be all the editor you need for many projects.

Sound Forge Audio Studio 8

This is Sony's $70 "light" version of Sound Forge. But it gives you similar functionality to the pro version with the exception of being limited to 16-bit audio files with sample rate up to 48 kHz and lacking the scrub function and volume envelopes. It lacks about half the processing tools and metering, so it's not as complete a "mastering" tool as the full version. But it will work fine to spiff up your field recordings. A plus, particularly if you're a beginner, is a good set of tutorials.

Wavelab Essential

This is Steinberg's $150 scaled-down version of Wavelab. It's based on Wavelab 4.0, offering both the Montage and Wave editing tools, CD-burning, ASIO and VST support, but sample rate is limited to 96 kHz. It's a good way to start out on the Steinberg path. If you like it and want the features of Wavelab 6 the upgrade is $400.

Audacity

Audacity is a remarkably complete recording and editing program available free at audacity.sourceforge.net. Audacity was developed by a group of enthusiastic volunteer programmers under Sourceforge's open source project, and includes many features of the commercial programs. It uses the cut-copy-paste model, and you can have as many windows open as needed to handle multiple source files. There are no automatic crossfades, similar to Sound Forge, but the program comes with a reasonable collection of tools and limited VST support. A nice bonus is that Audacity can play back and mix multiple files. This is handy in the production room when putting a music bed under narration, or even doing multitrack recording one track at a time.

— Mike Rivers
A plethora of new flash memory and minidisk recorders has hit the market, so there will be more stereo recordings coming in to your shop that never saw the inside of a DAW. And nearly all of these can benefit from some editing. You'll find that editing a stereo mix is somewhat different if your editing experience has been primarily in working with individual tracks in a DAW. It's much easier to identify edit points when you're looking at or playing a single instrument than a full mix. Any multitrack DAW program can edit a stereo file. But this round-up focuses on programs dedicated to two-track recording, playback, editing and processing.

Examined are several programs, which span the range from basic editing with a little tweaking to a full suite of mastering tools. All work with files in the WAV format, and most can import and export audio data in various data-compressed formats. So here's an overview to whet your appetite and give you some idea of what's available.

Sound Forge 8 – The Elder Statesman

Sound Forge ($299), originally developed by Sonic Foundry, started life as a sample editor back when Korg, Akai and Roland ruled the music sampler kingdom. It's now published by a division of Sony, and has grown from its modest beginning nearly 15 years ago into a very sophisticated program for working with stereo or mono recordings. Its sample editor heritage shows through in that it offers tools for generic loop editing, as well as creation and editing of loops for Sony's ACID music construction program.

Sound Forge uses the linear cut-copy-paste model for program editing. It crosses over to the source-destination model in that you can have multiple audio workspace windows open simultaneously, using one window as a destination in which to assemble chunks of material copied from the source window(s). If I'm working on a project with a single long file as the source — cleaning up a concert recording for broadcast, for example — I'll work on it in a single window. Most of the editing for a job like this consists of cuts, and it's not likely that I'll want to rearrange the order of the songs. If I have several source files — several nights of shows to be compiled "best-of" fashion into a single program, for example — I'll open each source in its own window so I can pick out the best pieces for the compilation.

There are about as many ways of working a project as there are tools, and Sound Forge has a lot of tools. I found that it's possible to make very precise edits in Sound Forge. But an efficient way of compiling a project is to grab a bit more than you know you'll need from the source file, slap it in to the destination window (remembering to drop a marker to remind you that you have some trimming to do) and then trim up this rough compilation after you have all the pieces in the proper order. Sound Forge allows you to preview the results of a cut before actually making it by playing a couple of seconds (configurable) before the cut, skipping over the cut and playing a couple of seconds beyond the cut. It's easy at this point to adjust the length and position of the cut if the pieces don't fit together smoothly, then to give another listen.

Sound Forge has an extensive set of keyboard commands, far too many to remember all (I lost count at around 160). But it's worth taking the time to learn some. It uses the conventional Windows shortcuts for frequent commands including "Cut," "Copy," "Paste," "File Save" and "File Open." Most shortcut keys are pretty well thought out as to how they're placed on the keyboard, allowing you to work with the mouse in one hand the other hand on the keyboard. Right-clicking with the mouse brings up a short context-sensitive list of commands, which usually includes what you want to do.

A handy tool, which can be a default or selected when desired, automatically makes a cut at the waveform zero crossing nearest to the selection boundaries. You can further specify (another preference) whether you want to cut on a positive- or negative-going slope, or the nearest zero crossing. The cursor can similarly be made to snap to a zero crossing. Zero-crossing splicing seems like a good idea (it's a bit of 'Net wisdom that isn't really true). Unless you're splicing two waveforms of the same frequency, however, there will be a change in slope at the splice point no matter where it occurs, and that can produce an audible click.

I consider it to be a serious omission that Sound Forge has no automatic or simple way to crossfade over a splice. It's possible that the developer believed the "zero crossing splice" theory and left it at that. There's a crossfade function, but this fades across the entire length of the selection, useful for fading music down when narration comes up, but not for making "clickless" splices. It's possible to crossfade a splice using several steps, but it's not something that you'd want to do 100 times in a program.

The mouse wheel can either scroll through the length of the file or zoom in and out. Zooming with the mouse wheel makes this a breeze when trying to locate an edit point in a large file. It turns out in practice, however, that scroll wheel control is fussy about the mouse, or at least the mouse driver. Zooming didn't work on one of my computers that has an off-brand mouse. The scrub tool is handy for "shuttling the tape" and listening to the monkey chatter, however, it doesn't come near to emulating the feel and preciseness of rocking tape reels. You still have to locate tight edit points visually or guess-and-by-gosh. ("By gosh, that sounds OK to me.")

Sound Forge 8, being a Sony product, can import and export the latest version of ATRAC compressed files used in Sony's new MZ-M200 Hi-MD recorder, as well as plain vanilla WAV files, MPEG-1 and –2, MP3, AIFF, OggVorbis and a bunch of formats I've never heard of. You can't edit video with Sound Forge, but it will import and export QuickTime, RealMedia and Windows Media files, displaying the video in frames above the audio track you're editing so you can see where to place cues. Audio — 8-, 16- and 24-bit — at sample rates from 2 - 192 kHz can be recorded, played, imported and exported. A resampling function allows you to convert sample rates up or down. There's a wealth of DirectX processing plug-ins bundled with the program, and it also supports third-party DirectX and VST plug-ins.

Sound Forge is fairly intuitive for basic operations, but it uses some unfamiliar terms for familiar items and functions weren't always on the menus where I expected to find them. The full manual, which I found difficult to use, is a PDF on disk. There's a printed Quick Start guide, which is a too-quick overview of the workspace and a rather sketchy reference to commands. I found myself getting befuddled quickly. There are plenty of worthwhile features here, but Sound Forge demands time to learn to use it effectively. I found myself looking through the printed Keyboard Commands reference to find what I was looking to do.

For more information visit www.sonycreativesoftware.com/products/soundforgefamily.asp.




Sony Sound Forge 9

Since the completion of this round-up, Sony Sound Forge has undergone a major update. Sound Forge in its new Version 9.0 offers 16 substantial updates: multichannel audio recording, file editing and processing; drag-and-drop between channels; hardware meters with output gain control; phase and mono-compatibility meters; wet/dry mix and crossfade options for effects; Channel Converter tool for multichannel audio; multichannel spectrum analysis; enhanced user interface; multichannel WMF support; 5.1 Dolby Digital AC-3 export; Gracenote MusicID; Noise Reduction 2.0; iZotope mastering effects bundle; QuickTime and MPEG-1 and 2 import; and VST support.

Fast Edit 4.0

Fast Edit, originally published in 1994 as Fast Eddie, is another old-timer that still has a large group of loyal supporters. Version 4.0 ($199, released in 2001) takes advantage of 32-bit Windows (95 and later), but has changed little in terms of operation since the initial release. It now handles sample rates up to 96 kHz and word lengths up to 24 bit, storing high-resolution files in a 32-bit format. Minnetonka Software told me that a new release is in the works. The name describes the program well: It's fast, both in execution and operation, and it edits.

Fast Edit is a classic source-destination editor. Open or record a file in the bottom read-only window, then paste segments (or the whole file) into the top window where clips can be trimmed, moved by cutting and pasting, and tweaked with fades and audio processing tools. Level adjustment, normalization, speed-and-pitch change, a two-band equalizer and the ever-so-'70s reverse play are included. Fast Edit additionally supports DirectX plug-ins, and with no coaching at all found all of the plug-ins that Sound Forge installed on my computer.

Having only one source window might seem skimpy if you're assembling a project from several sources. But once a source file has been opened and its graphic data generated even an hour-long audio file opens instantly. It's no big deal to re-load a source. Most navigation can be done with just the mouse once you learn the "chords." Clicking the right mouse button while holding the left button, for example, switches between the play and selection cursors; right-clicking plays a selection. There are toolbar buttons for copy, paste, selection, loop playback of the selected area and zoom in/out, plus keyboard shortcuts for everything.

Tabbing moves between markers, and the < and > keys move to the next or previous splice (a very handy feature). You can also open a "History" list and navigate through an edited file by clicking on list entries.

Enable automatic crossfading and a crossfade is performed across each splice or the boundary of each cut is faded in or out. Default fade length is 10 ms, which seems to work all the time; it can be easily adjusted, however. It's nearly impossible with crossfading to make a splice that clicks. Fast Edit doesn't offer volume envelopes, but you can create complex fades or volume changes by setting markers at volume change points and performing a multi-segment fade between them. The starting and ending volume of each segment can be set, and you have a choice of a linear or logarithmic gain change between markers.

Normally the edited project is saved as a WAV file, but it can also be saved as an "Edit List" file. This is convenient if you'll be coming back to the project for another editing session. The Edit List saves the location of any markers you may have set in addition to conserving disk space.

The manual is concise and a bit more tutorial than Sound Forge's. It's not quite up to date with this software version — all the functions are there, but, for example, screen shots for the recording parameters stop at 48 kHz and 16 bits. The manual, like Sound Forge's, contains a long list of keyboard shortcuts that you'll learn when you use them often enough. The program is very straightforward. Once you learn the distinction between edit and selection cursors, everything else is pretty obvious. Fast Edit isn't loaded with features, but what it does it does very well.

For more information visit www.minnetonkaaudio.com/fastedit/fastEdit.html.

Wavelab 6

Calling Wavelab 6 ($700) a two-track editing program is like calling a Ferrari a compact car. Yes, it is, but it's so much more. To do the program justice in a review would require more space than this whole article; so I'll focus on its editing capabilities and features that relate to editing.

There are two primary work spaces for audio editing. First is the Wave window, where drag-and-drop and cut-and-paste editing is performed. You can have as many of these open as you need for multiple source files, and freely drag selected areas from sources to one "master" destination Wave window.

The small upper panel provides an overview of the file, while the lower panel is where the work gets done. The zoom level and position of the two displays is independent, allowing you to use the overview to navigate through the file, bringing areas that need work into the main window. The mouse scroll wheel, like in Sound Forge, works as either a zoom or scroll, though they're opposites: Holding the Ctrl key in Sound Forge switches from zoom to scroll, while in Wavelab it switches from scroll to zoom. There's some intelligent interaction between the two windows. Selecting an area in the overview window sets the zoom level in the main window so that the selection just fills it.

It took me a while to find it in the Wave window (it's called "Smooth Delete"), but you can do a smooth crossfade over the joint when deleting a segment. Except to paste with a crossfade (something that's automatic in Fast Edit) you need to work in the Audio Montage Window.

The Audio Montage window is Wavelab's object oriented editor. The overview window shows the arrangement of clips (objects) rather than waveforms, but the overall appearance is similar to the Wave window. Sections can be pasted from Wave windows, or a large file can be loaded into the Montage window and split into objects that can be individually manipulated.

It's usually easy to locate split points visually, and this is usually the quickest way to divide a concert recording into individual songs. Split points need not be terribly accurate since you can always resize clips just by dragging the edges. There's an auto-split tool that will split at markers, split at "silence" (defined by parameter you can set), or split at specific time intervals. Clients for CDs of conference recordings often want them indexed at one- or five-minute intervals, even if that occurs right in the middle of a word.

You can perform all the standard editing functions in the working (lower) panel of the "Montage" window. Cuts and pastes can be made with crossfades, segments can be resized, crossfade times and overlaps can be adjusted, volumes can be matched, processing functions applied, and, when you're satisfied with your edits the file can be rendered to create a new WAV file with all the edits and tweaks.

The cursor in Wavelab represents a rather large toolbox, and what it does is a function of its vertical position. This takes some concentration and a steady hand to be sure of doing what you want. There are a lot of functions and a lot of menus in this program — it's one where you'll find your repertoire expanding as you use it and discover a new tool.

Wavelab offers a slew of power tools under its "Analysis" menu. Most are oriented toward mastering tasks, but there are some that will help to make your edits better, or make your job as an editor easier. One example is the "Spectral Editor." This displays time (location in the audio file) along the horizontal axis and frequency along the vertical axis. Amplitude is represented by color. Opening this display and applying the Spectral Editor helps you locate a bothersome frequency and remove it from the audio file without cutting out everything else that occurs at the same time. It's an alternate solution to simply cutting out the segment with the offending noise.

The Spectrum Editor is similar to processing a segment of the sound file with a narrow band pass filter (and this may indeed be what happens under the hood), but the value of the Spectrum Editor is the ease of visually locating trouble spot, and the analysis figures out what's required to remove or attenuate it.

It's particularly effective on two bugaboos of live recording: feedback and audience coughs. This illustration shows a burst of feedback marked for processing. Feedback is an easy-to-see example since it's usually, at least initially, a single frequency, that appears as a horizontal line in the spectral view. Note its color change from blue to yellow and back as the squeak builds and drops in intensity. A keystroke removed it without noticeably changing the underlying speech.

The "Average Loudness" analysis tool displays the apparent loudness over the selected area (or entire file). I found it to be a useful aid in matching levels when replacing a section with the equivalent piece of an alternate take. I could determine if a volume adjustment was required by analyzing the loudness of the original segment and its replacement, and, if so, how much of an adjustment. It's a timesaver, and I've found that using the analyzer's numbers works the first time just about every time.

Normally I consider "normalization" to be a four-letter word. But Wavelab's "Loudness Normalizer" is interesting, though its effect can be quite comical if applied to the extreme. The Loudness Normalizer in better hands than mine when used by a skilled mastering engineer is effective for bringing levels into the ballpark when assembling a CD from various sources or evening out multiple speakers in a conference or forum recording.

Wavelab is a hugely powerful program, and this quick overview barely scratches the surface. It can create and burn DVD-Audio discs as well as audio CDs, and it has a multitrack mode for editing surround productions in up to 7.1 format. VST and DirectX effects are supported, and a tasty assortment is supplied with the program. It may be more than you need if you're not planning to go into the mastering business, and harder to learn for routine editing tasks than other programs. But it's a good choice if you want to have a wealth of tools at your fingertips.

For more information visit www.steinberg.net.