Digital Audio Editing

Correcting and Enhancing Audio in Pro Tools, Logic Pro, Cubase, and Studio One

Simon Langford

• Editorial: Routledge
• Año: 2013
• ISBN:
  9781134111374
• Edición: 1
• Categoría: Mezzi di comunicazione e arti performative
• Categoría: Teoria e gusto musicale

Whether you're comping a vocal track, restoring an old recording, working with dialogue or sound effects for film, or imposing your own vision with mash-ups or remixes, audio editing is a key skill to successful sound production. Digital Audio Editing gives you the techniques, from the simplest corrective editing like cutting, copying, and pasting to more complex creative editing, such as beat mapping and time-stretching. You'll be able to avoid unnatural-sounding pitch correction and understand the potential pitfalls you face when restoring classic tracks. Author Simon Langford invites you to see editing with his wide-angle view, putting this skill into a broad context that will inform your choices even as you more skillfully manipulate sound. Focusing on techniques applicable to any digital audio workstation, it includes break-outs giving specific keystrokes and instruction in Avid's Pro Tools, Apple's Logic Pro, Steinberg's Cubase, and PreSonus's Studio One. The companion websites includes tutorials in all four software packages to help you immediately apply the broad skills from the book.

CHAPTER 1 Audio Editing 101

DOI: 10.4324/9780203512890-2

Editing History: Tape

It was only with the advent of multi-track tape recorders that any kind of meaningful editing became possible. Audio editing (in the very loosest definition of the term) was technically possible with earlier mono or stereo tape recordings, but any editing would have been limited to very simple repeating of sections or removal of sections of the whole recording. When multi-track tape recorders came into the world (courtesy of a company called Ampex in the 1950s) a few more options became available to the creative (and patient) engineer. On the surface there doesn’t seem to be any more flexibility, because editing of a multi-track tape in the way it had previously been done still involved cutting sections of the entire tape (across all the tracks at once) to do the edits. The advantage with a multi-track tape is that individual tracks could be copied to another tape and then copied back over after the edits had been done—so it was time-consuming, but definitely possible.

With the required patience, it was possible to copy sections from other places within the track and paste them, although that would have been accomplished by recording the sound or section in question out onto another tape machine and then rerecording it back onto the original multi-track tape at the appropriate place. This was far from easy, and, even though the two tape machines could possibly have been synchronized through the use of a time code recorded on to each of them, it would still have been a very difficult process, and there certainly wouldn’t have been the sub-millisecond accuracy that we have now. Nonetheless, it was the first of many steps toward where we are today.

In addition to the copy and paste abilities of multi-track tape, there was also the option to process individual tracks or sounds using a variety of studio equipment. One of the corrective aspects of audio editing involves level and dynamics control. Multi-track tape recording allowed the engineers and producers to have control over the relative balance of the tracks after they had been recorded for the first time; in addition, individual tracks could be processed with compressors and other outboard equipment to allow for changes to the dynamics as well as level control. Things that we take massively for granted these days and can accomplish in seconds with the flick of a wrist and a few well-placed mouse clicks may well have taken many minutes, if not more back, in the early days of multi-track recording. This kind of level control could be seen as a part of mixing or production, but there are times when it will fall under the remit of the editor, so it was worth mentioning here.

“Comping” (short for compiling) became a realistic possibility as well, because a performer could record multiple takes onto adjacent tracks, and then the best sections of each performance could be noted down and the relevant tracks soloed (or channel faders on a mixing desk moved) at the appropriate time, and the resulting compiled track could be recorded on to another track on the tape—often called bouncing—or perhaps even another tape machine. Because the timing of the changes wasn’t always accurate, it is most likely that the changes between takes would have happened between lines of a song or perhaps, at best, between words. Nowadays we can quite easily edit between different takes, even in the middle of a single word or note and, with a little work, make the resulting edit sound utterly smooth and natural.

A further development of this concept of comping is in the ability to change individual drum hits within a recording if the drums have been recorded with a multiple microphone setup. There are many limitations to this, which we will look at in detail later on, but, with care, it is possible to remove an errant drum hit within the context of the whole recording, which simply wouldn’t be possible with a mono or stereo “mixed” drum-kit recording. Obviously this became possible only with the advent of multi-track tape (and in particular multi-track tape machines with a good number of tracks available) and is something that could prove very useful in fine-tuning a recording.

Looking at this very short summary of the editing possibilities of multi-track tape, it becomes clear that a lot of what we would consider to be the “corrective” editing processes that we use today could, in fact, be done with the medium of multi-track tape. What should also be clear, however, is that many of these processes were more of an art (and not always a 100% accurate one) rather than a science, and a great deal of skill would be required in order to achieve decent results, and, even if that skill level was attained, it was still a time-consuming process. The next stage of evolution of recording (and therefore editing) came with the advent of digital audio, and it was with this development that things started to become more accurate, reliable, and repeatable, and required less technical skill and time.

Editing History: Digital Tape and Hard Disk–Based

The idea of recording sound digitally actually goes back much earlier than most people think. The very first digital audio recording experiments can be traced back as far as the late 1960s and early 1970s, which predates the 1981 introduction of the CD format by over a decade. Just a year after the unveiling of the CD, Sony debuted its DASH digital multi-track systems, which offered twenty-four or even forty-eight tracks of digital audio recorded onto half-inch wide tape. These systems were, naturally, staggeringly expensive but were totally ground-breaking and paved the way for the systems that would follow later and at a much more affordable price. In use, these Digital Audio Stationary Head (DASH) systems (also available from Studer and Teac Audio Systems Corporation America [TASCAM]) were very much like a conventional analog tape–based multi-track in that, through the use of some very clever encoding and error correction, the tape could be physically edited and spliced in the same way that people were used to doing with analog systems. In fact, that was pretty much the only way you could edit with these systems. So in this regard they served as a very easy transition into digital recording. Most of the editing techniques that editors already used could still be applied. The only thing that was different was that it was a digital signal being recorded rather than an analog one.

One fundamental advantage of digital recording is that there is a much lower noise level on recorded signal, which means that multiple tracks layered up would be a great deal cleaner and clearer than a comparable analog recording. Another advantage is that, in the event that a digital track needed to be recorded or bounced to another track or another recorder altogether, assuming that the signal was kept digital during the process and not converted back to analog and then to digital again, there would be no loss of quality and no increase in background noise levels. One final advantage that is worth mentioning, and which could prove very important if trying to copy and paste sounds from different parts of the recording, is the fact that digital recordings have no inherent “wow” or “flutter.” Both of these terms relate to minor (or sometimes major) fluctuations in the speed of the tape as the recording is being made or played back. This is problematic, because an analog audio recording on tape has a direct relationship between the pitch of the sound and the speed of the tape. When there are minor variations in the speed of the tape during recording or playback, the pitch of the recording will vary with the changes in tape speed, resulting in a “wobbling” effect. With a digital system, this simply doesn’t apply. You can still change the pitch and tempo of a digital recording by adjusting the playback frequency to something different than the sampling frequency of the recording, but at least every part of every track will be affected by the same amount (and there will be no wow or flutter) which allows for much more consistency when cutting, pasting, and moving audio among different parts of the same recording.

It would take a different format entirely to start the transition to what we know today, and that came in the form of using hard drives as the recording medium. Although this is often thought of as a development that came after digital multi-track tape, the very first hard-disk multi-track recorder was actually a software add-on for the NED Synclavier II system and was released in 1982, the same year that the Sony DASH system was introduced. Because of the cost and scarcity of this system, tape-based digital systems were the ones that really pushed forward the digital revolution in multi-track recording. Though behind the scenes the work continued, and then, in 1993, the Otari RADAR (Random Access Digital Audio Recorder) was introduced. The key to its success, and the subsequent move to hard-disk recording, lies in the name: Random Access.

This one aspect of the RADAR system would become a huge selling point and would revolutionize the way that audio recording and editing (more so the latter) were carried out. Tape-based systems are essentially linear recording systems in both input and output. Sounds are recorded from a start point to an end point and then are played back (physical splicing and editing aside) from that same start point to the same end point. Hard-disk recording as offered by the RADAR system was slightly different. Of course the recording would still be linear in input—a recording would start at a particular point and continue on for a set amount of time—but the linearity of the output wasn’t required any more. This non-linear editing process allowed editors sample-accurate editing and the ability to select, cut, copy, paste, and move sections of audio without any physical process at all, and, best of all, this was the start of the ability to undo the edits if they either weren’t done correctly or simply didn’t have the desired effect.

The RADAR system had many limitations, though. While it could record up to twenty-four tracks of audio, the internal 1 GB hard drive only allowed for a total of roughly two hundred track minutes of recording (total number of minutes available over the total number of tracks) or eight minutes of twenty-four-track recording. This in itself wouldn’t have been a major shock, because a typical reel of tape running at thirty inches per second would give around fifteen minutes of recording. The problem was the cost of hard drives relative to tape. A reel of two-inch tape was by no means cheap, but, in 1993, a 1 GB hard drive would have cost close to $1000, which would have been a very limiting factor. It was, though, only a matter of time before hard-drive prices began to fall to more reasonable levels; at the same time, capacities start to increase to the point where a whole drive could record much more than eight minutes.

Even though these systems were inherently more flexible than tape-based systems, there were huge costs associated with them, so tape-based digital recording wasn’t quite dead just yet. The next big step in the evolution of tape-based digital multi-track recording came in 1992 with the release of the original Alesis ADAT recorder. This was a system that, at a relatively affordable price, offered eight tracks of digital recording (multiple units could be synchronized to provide up to 128 tracks if required) on S-VHS format cassettes. This may well have been an important factor in the success of the original ADAT machines. Open reel tape (as used on analog and DASH systems) could be very expensive, as well as taking up a lot of space to store. The S-VHS cassette format made the recording media for the ADAT compact and affordable, very much like the machine itself. It wasn’t all positives, though, as the splicing and editing ability that both analog multi-track machines and DASH machines possessed was now gone. Given that the tape was enclosed within the cassette and the cassette itself inserted into the recorder, there was simply no way of editing in the way that people had been used to before, even if the underlying technology (as in the DASH systems) had allowed it.

Hard-disk recorders were dependent on storage space, and, in the early 1990s, this was, as we have seen, still prohibitively expensive. However, as the various storage media prices fell, the number of machines on the market increased. By the turn of the century, there were eight-, sixteen-, and twenty-four-track hard-disk recorders on the market from many of the companies familiar to studio owners, such as Akai, Fostex, Roland, TASCAM, Yamaha, and many others at prices that were far more affordable. As these “all-in-one” digital studios started to become more accessible, the ADAT/sampler setup became less attractive, and eventually the ADAT recorders were discontinued, and the move to nonlinear systems was pretty much completed. However, while these stand-alone recorders were being developed, the same technology was being harnessed in a different way to allow software sequencers running on PC or Macintosh computers to add audio-recording abilities to their already established MIDI recording, and, when this became possible, the digital audio workstation, or DAW, was born.

Editing History: Daw

Computer-based digital audio recording, and the later change into what we now call DAW software, was a total game-changer for both the recording and editing processes. By not using any form of tape media for the actual recording, DAW software promised far greater track counts (limited only by computing power which, even at the time that DAW software first started to emerge, was already increasing exponentially), almost stupidly easy cutting and pasting, and significantly greater editing options.

Strictly speaking, the very first digital audio workstation was released in 1978 by a company called Soundstream, used hard disks (very low in capacity by today’s standards) for storage, and allowed for very basic editing of the recorded audio in addition to mix-down and cross-fades. The facilities of the system would be considered arcane today, but at the time they represented a quantum leap in audio recording and editing freedom. Most PCs of the era were text-based (DOS) systems, and this factor alone contributed to DAWs’ not really developing that quickly. Aside from the limitations of the technology (processing power and storage), these early systems were not overly user-friendly. You could argue that they were, in many ways, more user-friendly than working with tape, and there is some logic in that, but any systems that relied on text and numbers to define absolute positions and commands was in some ways counterintuitive to the natural and organic nature of music-making. What was needed was a better way of interacting with the underlying technological ideas of the Soundstream system.

That better way was a natural development of the move to PCs, which operated with graphical user interfaces (or GUIs). The fact that things were represented onscreen with objects and icons that were visually related to the things they represented allowed for a much more natural way to operate computers. By the late 1980s, there were many affordable computer platforms using these GUI operating systems, which had already had sequencing software packages written for them to allow the control and recording of MIDI instruments. The most common platforms in use for music sequencing at the time were the Apple Macintosh, Atari ST, Commodore, Amiga, and generic “PC.” The processing power of all these machines at the time was starting to make audio recording a realistic proposition, and in 1989 a company called Digidesign released their ground-breaking SoundTools software. While this would later evolve into the now almost ubiquitous Pro Tools, this early relative was limited to only mono or stereo recording and was less of a DAW than a stereo digital recorder with some quite advanced editing features.

One of the most interesting was the FFT Window, which provided a Fast Fourier Transform view of the audio recording. This provides a three-dimensional overview of a sound that is similar to that shown by a spectrum analyzer (with frequency on the horizontal axis and amplitude on the vertical axis) but with the additional third dimension showing the change of this spectrum over time. If you imagine a series of snapshots of a spectrum analyzer display stacked up one behind the other, then that is the basic concept. This FFT view allows a more intuitive view of how a sound changes over time, but, at this period in DAW evolution, while the FFT display was a valuable addition for figuring out what needed doing, it was purely an analytical tool, and no changes could be made directly to the sound in this view. The FFT view would be used to figure out what needed to be done and where, and then the traditional tools would be used to actually make the changes. Nonetheless, it paved the way for more-advanced editing techniques later on.

It didn’t take long for the technology and ideas used in SoundTools to make its way into more advanced software, and in 1990 a company called OSC (distributed by Digidesign) launched their Deck software. This was similar in functionality and principles to SoundTools but allowed four simultaneous tracks instead of the two that SoundTools was limited to. Digidesign were already well aware of the potential of the Deck software and actually licensed the core technology for use in the first version of their Pro Tools software that was released 1991. This first Pro Tools version was still limited to eight tracks, but, like so many developments in digital audio, the limitations were entirely due to the hardware that the systems were running on. Everybody realized that it wouldn’t be too long before eight tracks became a realistic possibility, once the computing power and hard drive speed allowed. So Pro Tools evolved and allowed greater numbers of tracks over the years, and eventually became a very good alternative to a multi-track tape system. Later versions added in MIDI sequencing facilities as well to provide more of a one-stop studio environment. And, of course, as the processing power increased, the editing options and technologies increased and diverged to include “plug-ins” that allowed third parties to create mini software applications that ran inside of Pro Tools.

Meanwhile other companies were heading toward the same solution but from the completely opposite direction. There were two software companies that many considered to be the dominant and industry-standard ones when it came to MIDI sequencing software: Steinberg and C-Lab/Emagic. Both companies had developed software that had, over a number of revisions, become very sophisticated, and in use mirrored many of the paradigms of multi-track tape recording, only using MIDI instruments instead of audio recordings and, in doing so, retaining a greater degree of flexibility and editing freedom. But these applications lacked the ability to include any real audio recordings unless the user was prepared (and able to) record into a MIDI sampler and use the sampler effectively as an audio recording “add on.” Many did just that, but it still wasn’t an ideal solution, so work continued to bring native audio recording and editing functionality to the flagship products. And then, in 1991, Steinberg introduced Cubase Audio for the Macintosh (audio-capable versions of Cubase didn’t arrive on the Atari platform until 1993 and on the PC platform until 1996, owing to the relative limitations of those platforms compared to the DSP-enhanced possibilities of the Macintosh platform), which was a real game-changer. For the first time there was software that incorporated all the flexibility of MIDI sequencers with the options and additional benefits of audio recording and editing. In this respect, at least, it put Steinberg well ahead of the game, as at this time Pro Tools was strictly audio-only and Logic (Emagic’s direct Cubase competitor) was still MIDI only. It wasn’t until 1994 that Emagic would release a version of Logic (again for the Macintosh platform first) that would include audio recording features. Another manufacturer of historical significance is Mark of The Unicorn (often known simply as MOTU), who released their Digital Performer in 1990. This was originally created as a front-end to Digidesign’s Audiomedia hard-disk audio-recording platform, which added sequencing abilities to the audio features already available in Audiomedia. In doing so they added another possibility to the already rapidly growing array of options.

In addition to the extensive feature sets of these newly developed packages, there were continued advances in the quality and feature sets of two-track audio editing software. Fundamentally these offer many—if not all—of the features of the audio side of a comparable DAW but work on a single two-track (stereo) audio file at any given time. If you think of it as working with a two-track “master” audio tape or a single track of a DAW, then, conceptually, at least, you will be in the right ballpark. Additionally, they often have features not available in DAWs, as ...