10	The Audio Tools

 

OK, so not everything is going to be perfect after your recording session. Sometimes people will speak too loudly or softly, or worse, they will do both during the same recordings. There will be some noise, no matter what you do. And frankly, you’re going to want to change things that aren’t technically bad audio, just because, well, because you want them to sound differently. Of course, you should always strive to record the highest quality audio you can while shooting, but this isn’t always within your control.

FCP has an excellent suite of new tools for addressing audio issues. Similar to opacity, audio levels can be adjusted and keyframed on the sequence as well as in the Audio tab of the Viewer, and from a special new tool in the Toolbench. FCP also has Audio Filters that can be used to clean up a number of common issues. Finally, there are some pretty cool tools that make it easy to get quick, classy voiceovers into your productions.

Fixing Audio Levels

First, we’ll take a look at the basic audio tools in FCP. In your project, open up a new sequence and edit a clip into it. Select the clip and go to Modify > Make Stereo Pair (keyboard shortcut, Option-l), if they are not already linked as such (see Figure 10-1). In the Log and Capture window you can specify that FCP captures either Stereo Pairs or Mono channels. Look on the Capture Settings tab for the Input Settings pulldown on the lower left side of the window. How you set this tab determines how the audio will be linked after capture.

To understand the difference between a stereo pair of audio clips and a couple of mono clips that are simply linked together, you have to understand the difference between mono and stereo sound. Mono sound is unidirectional. In simple terms, this means that the sound you hear is flat and appears to be generated from all directions at once.

Stereo, on the other hand, uses two speakers to simulate depth, and thus position (or origin) in the sound field. Sounds emitted by stereo speakers, if mixed properly, do not appear to originate directly from the speakers, but rather somewhere between them. It is this variance of levels of a single sound to two different speakers, or ‘panning/stereo mixing’ that gives stereo its depth, realism, and presence.

Figure 10-1 To make any clip linked Stereo, select the clip and go Modify > Make Stereo Pair.

When you make two audio clips stereo in FCP, as we have just done, you add this mixing option. But even if you don’t intend to do any heavy mixing, making tracks stereo can benefit you with a special sort of convenience when setting and adjusting audio levels, ince it is assumed that you want to increase stereo channel levels equally, when you make two audio tracks stereo, any level adjustment you make to one is automatically applied to the other. A stereo pair will appear on a single audio tab in the Viewer (see Figure 10-2). Any Audio filters are automatically applied to both audio clips. In the sequence, two little green triangles appear at the head of the audio clip to indicate that these two audio clips are linked in a special way. Even if you unlink your clips using the Toggle Linking button in the Timeline window, the two clips will still be locked together in this special elationship.

After selecting the new stereo-ized clip, double-click to load it into the Viewer window. Click on the Audio tab and a window opens that contains two wavy, jagged lines. These are called waveforms. A waveform is essentially a graphic representation of the volume levels of the audio in the clip. Where you see spikes in the line, the audio level is higher; where the jaggies are shorter, the levels are lower.

These levels correspond to the levels you may have already seen in the clip in the sequence when waveforms are toggled on (keyboard shortcut, Command-Option-w). They are also seen in the Audio Level Meter, a small thin window that normally resides next to the Toolbar, when the windows are laid out in the default arrangement. If you hit the spacebar to begin playback, you will hear sound and see the level meter registering the rise and fall of the audio levels.

Of course, we are not limited to the levels that we have when we capture. If you look near the top of the tab, you will see a slider marked Level. This red slider corresponds to the red horizontal level line down in the waveform area of the window. To change levels,

Figure 10-2 Stereo linked clips use one single tab on the Viewer, and filters applied to one audio clip are automatically applied to both left and right channels.

you can either move the slider left or right, or you can use the mouse pointer to move the level line in the waveform tracks up and down (see Figure 10-3). Use the mouse pointer to grab the red line in the waveform track and drag it up. As you do, you will see a small box appear with a number and “dB.” The dB stands for decibel, a unit of measurement for force which is often used to describe sound or audio levels. As you drag, you’ll see the slider follow it. When you see “8 dB,” let go.

This audio level rubberband line can also be found on the sequence. In the Timeline window, enable the Clip Overlay button. The clip’s audio tracks will display a similar level line. If you grab that line with the mouse pointer, you will see that it is also at the 8 dB level.

Figure 10-3 On the Audio tab, you can boost or cut levels with the slider or on the rubberband line in the track.

If you drag it back down to 0 dB, you will see the lines in the Audio tab in the Viewer window follow it.

Keyframing Audio Levels

Of course, we are not limited to just raising or lowering the levels of an entire audio clip. We have already learned how to keyframe opacity levels in the sequence. We’ll use the Pen tool again to change the audio levels. With both Overlays and Waveforms turned on, select the Pen tool (keyboard shortcut, p), then proceed to the audio clips in the sequence. Position the mouse pointer over the rubberband line in the audio tracks, and click to create the first point. Click elsewhere, either prior to or after this first point, to set a second keyframe, place the cursor between the two keyframes, and drag up or down. As you do you will also see the little “dB” box appear informing you how much you are raising or lowering audio levels.

Also notice that a keyframe can be dragged and repositioned on the Timeline (left or right). When you change the position of the keyframe in the sequence, the line between the two points extends, and just like with opacity keyframing, the transition between the two audio level values takes longer. So, for instance, if you wanted to slowly increase an audio track from quiet to loud over four seconds, you would set the first keyframe at a relatively low minus dB value at the first frame where the audio starts. Then set the last keyframe at the fourth second at a higher, more suitable, audio level (see Figure 10-4).

The Audio Mixer Tool

New to FCP 4 is yet another window in the Tool Bench. This is perhaps the most eagerly-awaited addition to the FCP toolset, the Audio Mixer. The Audio Mixer is a full-fledged mixing interface, complete with track and master faders, solo, muting, and panning controls. Unlike the previous version’s audio mixing toolset, the Audio Mixer is built to

Figure 10-4 Just as with opacity keyframing, audio levels can be boosted or cut directly on the Timeline.

allow you to mix your audio tracks on the fly, raising and lowering audio levels for each track independently as you playback. Nothing could be faster.

Go to Tools > Audio Mixer. When you do, the Audio Mixer will appear in the Toolbench (see Figure 10-5). To the far left of the window you will see all the tracks in the current sequence listed. Next to this, on the right, are the fader controls for the tracks. Above each fader are the Pan, Solo, and Mute buttons. Below the level faders are the readouts for the level of the fader expressed by the fader.

Load a new video clip with two audio clips and edit it into the sequence. Edit more audio clips to the sequence in more audio tracks until you have at least six audio tracks in the sequence and Audio Mix window. If you need to add more tracks to the sequence to accommodate the merged clip in the Viewer, go Sequence > Insert Tracks and add additional ones. Notice that as you add tracks to the sequence, faders are added to the Audio Mixer tab. Fader channels, as tracks are sometimes referred to, are linked to sequence tracks (see igure 10-6).

Hit the spacebar to initiate playback. Then, grab any fader you like and adjust. You will hear audio levels drop for the contents of that track, and its stereo counterpart if there is one (you may want to hit Control-l, the keyboard shortcut for Loop Playback, so that the

Figure 10-5 Go Tools > Audio Mixer and look to the Toolbench for the Audio Mixer.

Figure 10-6 Adding tracks to the sequence adds faders to the Audio Mixer window.

clip repeats itself enough times for you to get the hang of the faders). Notice that as you adjust it, it changes levels on the fly to match, without stopping playback. This is a major advantage of using faders for setting levels.

You will notice, however, that when you adjust the levels, they affect the entire clip, rather than setting keyframes for level within the clip, as would be most useful to us in cleaning up lousy tracks. This is for your own safety, though. You can set levels directly from the faders, but you have to enable it with a toggle switch. This is to keep you from accidentally wiping out levels you already set. Look in the top right corner of the Audio Mixer window for the button bar (see Figure 10-7). The Record Keyframe Audio button is preloaded there, because the FCP team knew that you would need to access it while in this window pretty frequently.

When you want to keyframe audio levels for your tracks, click the Record Keyframe Audio button. Note that this overides the Clip Overlays rubberbanding tool in the Timeline window. It doesn’t matter whether the Clip Overlays toggle in the Timeline is enabled or not; when the Keyframe Audio button is engaged, you are setting audio keyframes with every change you make in the faders. It is in your best interest to have the Clip Overlays toggle on in such circumstances, because that is the only way you can actually see the keyframes you are creating (see Figure 10-8)!

With both Record Keyframe Audio and Clip Overlays buttons enabled, start playback in the sequence. As it plays back, adjust various faders. After a couple of loops, hit the spacebar to stop. Look into the sequence and the clip has new keyframes for each track you adjusted with the fader.

Depending on how you have your User Preferences set, your keyframes may look similar or very different to what you see in the images with this section. That’s because FCP lets you determine the precision of the keyframes as you change them. This can make a big difference both while you are setting them and adjusting them after the fact. Go to Final

Figure 10-7 The Record Keyframe Audio button must be enabled to create keyframes in the sequence on the fly using the Audio Mixer.

Figure 10-8 Make sure you keep the Clip Overlays toggle enabled so that you can actually see the keyframes you create on the fly with the Audio Mixer faders.

Cut Pro > User Preferences and look on the General tab. Under the Audio Playback Quality, you will see a toggle switch and pulldown tab for “Record Audio Keyframes” (see Figure 10-9). The toggle switch is the same function you enabled from the button bar in the Audio Mixer tab, so you need not deal with that here.

The pulldown menu is a different story, though. The pulldown gives you three options: All, Reduced, and Peaks Only. All speaks for itself. When you make a change in the fader for the track, it record keyframes as fast as possible, getting pretty close to a keyframe per frame of video. Reduced records keyframes, but less frequently. The keyframes are still what you are inputting with the mouse, but are sampled less frequently than with All. Peaks Only records only the most extreme adjustments you make with the faders, thus, generating potentially the fewest number of keyframes to edit later.

In practice, using All effectively is nearly impossible, unless your audio tracks are so screwed up that you would have to use an unlimited number of keyframes to clean them up anyway! Peaks Only is very difficult as well, since this depends on audio tracks that don’t really need much adjustment. Reduced is usually the best choice. It give you plenty of minute control, and don’t forget that you can always add extra keyframes manually with the Pen tool.

Figure 10-9 The toggle, plus a variable control for setting the precision and frequency of the keyframes you create using the Audio Mixer, is set in the User Preferences General tab.

Another great piece of functionality in the Audio Mixer tool is the View: Track Display ets (see Figure 10-10). In the top left corner of the Audio Mixer tab, look for “View:” and four buttons labeled 1, 2, 3, and 4. These buttons let you program display sets of faders. This means that at the push of a button, you can show certain tracks and hide others that are commonly adjusted in common.

To use this function, you just hide or show tracks while a view set is engaged. When ou start FCP, the first set “1” is engaged, with the default of having all tracks displayed. We will leave this as is, so that we can always switch to a view of all tracks simultaneously. Click on “2.” Initially, this track also has all tracks displayed. Move the mouse pointer over to the left side of the window and click A1 and A2 such that the accompanying circle is white. When you do, the A1 and A2 tracks disappear from the fader columns. witch back to “View: 1” and you will see the faders reappear.

“View: 2” now has a different display set from “View: 1.” You can continue to customize the view sets for any of the buttons. This tool is most useful for isolating views based on the audio elements you will encounter during video editing situations. You’ll commonly put your synced audio tracks (from the camera recording) and dialogue on A1 and A2. Music beds, sound tracks, and other additions might be on A3, A4, A5, and A6. Other elements like room tone and sound effects might be dedicated to A7 and A8. But, rather than always look at the huge range of tracks simultaneously, put together display sets so that you can mix your dialogue and sync tracks without dealing with 8 faders at once.

Each track has Mute and Solo buttons, located above the fader (see Figure 10-11). These buttons have counterparts on the sequence that we saw earlier in Chapter 6.

Figure 10-10 Track Display Sets let you customize which tracks are visible in the Audio Mixer, and let you switch between such customized views with a single click.

Figure 10-11 There are Mute and Solo buttons that correspond to the Mute and Solo buttons in the sequence tracks; each enables the other.

The Mute button, identified by the speaker, makes the track temporarily inaudible. The olo button, symbolized by the headphones, does the opposite. When you hit the Solo button for the track, not only does its Solo button change color, but all other tracks are muted. Notice that if you hit Mute or Solo, the corresponding button is enabled in the sequence.

Finally, the Pan slider is also keyframeable. Panning, as mentioned before, generally efers to stereo mixing. Stereo tracks give presence and dimensionality in sound by adding levels of a particular sound element to one speaker and taking it away from another. Although this has to do with stereo mixing (since you can’t pan in mono, stereo, by definition, equires two sides to pan between), you can still set panning keyframes, because panning is ultimately interpreted by the audio output of your FCP system.

If, for instance, you are using eight mono tracks with a Firewire DV system, your eight tracks are being mixed down by default to two stereo tracks for Firewire output. The Audio Output tab of the User Preferences allows you to create a preset for how your audio leaves FCP’s Timeline and enters the real world. If you have hardware support other than Firewire DV, you can actually keep your tracks discrete to multiple channels if you wish, rather than mix them into stereo mixdowns.

Take a look at the master control on the right side of the Audio Mixer tab (see Figure 10-12). This master control has a fader and a level meter, as well as a master Mute. Each track in the Audio Output Preset you are using is represented next to the fader. If you are using stereo two-track mixdown in the Audio Output tab, the default for those using Firewire DV, then you will only see two channels in the meter. These meters display the tracks of audio leaving FCP. If you were, on the other hand, using a MOTU 828 Firewire Audio digitizer which supports eight channels of input and output, you could set up an Audio Output preset for either four stereo pairs or eight discrete mono audio outputs. If I enabled this preset, the master control level meters will show all eight output tracks, even if there is nothing in those tracks, and even if there are fewer than eight tracks in the sequence!

Figure 10-12 The Master Fader, all the way to the right of the window, will have an output track for each output track you have in your current Audio Output Preset from the Audio Output tab of the User Preferences.

Just above the master fader is a downmix button in the Audio Output Preference. Even if you have access to a multi-channel output, such as the aforementioned MOTU solution, much of the time you will likely not be mixing that multi-channel audio. You will, in most cases, be quite happy to cut with two channels of mixed-down stereo. Instead of creating new sequences that use a separate Audio Output preset, you can simply click this button to automatically mixdown all tracks into two stereo tracks. Odd-numbered tracks are mixed into the left channel, while even-numbered ones are mixed into the right.

It should be noted that if the Audio/Video Settings Sequence preset you are using does not have enough outputs for the multi-channel Audio Output preset you choose, FCP will ask you if you want to mixdown as described above, or simply to run with all channels above the first two tracks muted.

Using the Audio Filters

Audio filters are very similar to video filters in that you apply them to an entire clip. These filters are used to correct audio problems or add effects. When you apply the filter to the clip, its controls appear in the Filter’s tab in the clip’s Viewer. There are a lot of filters to choose from, each with its own particular function. We’ll look at a few of the more important ones and figure out what they do and how to set them for your individual clip’s needs.

The Three Band EQ

An equalizer is an important tool in cleaning up audio. To understand what it does, ou have to know a little bit about audio itself. Audio and sound are composed of two important features that make them “sound” the way they do. The first, amplitude, describes olume levels as we have been adjusting them using rubberbands. Amplitude describes how powerful, or loud, a sound is. In the waveforms we’ve been looking at, the higher the spike in the wave, the louder the sound. This is the amplitude of the sound or audio wave.

The second is the frequency. Frequency is the number of waves that happen over a period of time. Don’t confuse this with the sample rate, which has to do with how frequently your camera samples the audio waves. Frequency here means how many different waves happen per second in the sound you are hearing. Higher frequencies result in higher pitch sounds like whines, and lower frequencies result in lower pitch sounds like rumbles. The combination of frequency and amplitude determines how loud and what pitch the sound is.

Knowing this gives us a little leverage for fixing sounds. The EQ, or equalizer, is a tool that allows you to adjust the audio levels of a clip, but instead of adjusting all frequencies at the same rate, it allows you to change the loudness of individual frequencies in the sound, or instance, you may have audio in which the high pitched sounds are where all the noise is, but you want the middle pitch, or mid-range, and the lows to remain the same. With an EQ, you would simply lower the audio levels for the high pitch noises, while leaving the mid-range and low pitch sounds untouched.

The number of bands an EQ has determines the precision that the EQ has in addressing ranges of frequency. A twelve band EQ, for instance, will split up the whole range of human hearing (20 Hz to 20 KHz) into twelve sections, allowing you to effectively have four different bands of EQ for all three (high, mid, and low) ranges. That’s a lot of precision. There are even 24 and 36 band EQs that can isolate very small bands of frequency without touching others. Of course, with that level of precision comes a big price tag.

The very lowest number of bands you are going to encounter in an EQ is going to be the Three Band EQ, simply because this is the fewest number of bands that can be effective when separating frequencies. To apply the Three Band EQ, edit a fresh clip with one video and two stereo audio tracks into your sequence, and then double-click to load it into the Viewer window. Before you go any further, you need to turn off Record Keyframing Audio, which can be done either from the Audio Mixer tool button bar or on the General tab in the User Preferences. Otherwise, as you experiment with settings, you will be laying keyframes all over your clip!

Go to the Effects tab and look in the Audio Filters bin (see Figure 10-13). Inside are two more bins. FCP 4 ships not only with the set of filters FCP editors are familiar with from past versions, but also another set of filters in a bin named Apple. The Apple AUGraphicEQ filter, in fact, is an extremely precise EQ just like the ones described above. It covers 31 separate bands of frequency. We won’t use it for this exercise, because 31 bands is pretty difficult to use if you are still learning where noise frequently occurs in audio for video. The limited capability of the 3 Band EQ in the Final Cut Pro bin makes it a little easier to see what is going on when you EQ. Other than the number of bands, the functionality is the same. When you get the sense of how EQ’ing works, you can put the AUGraphicEQ’s 31 bands to use.

To apply the 3 Band EQ audio filter, look in the Final Cut Pro bin. The first audio filter in the list is the 3 Band EQ. Grab it and drag it over onto the audio tracks of the clip in the sequence (see Figure 10-14). If the filter only appears to be settling on one of the two audio clips, stop, select the clip, and go Modify > Make Stereo. If you only apply the filter to one of the two audio clips, you will only be EQ’ing half the sound! And if you apply the filter to the two audio clips separately, you will have to put in the settings for the filters for both clips separately as well.

Figure 10-13 In the Effects tab, you will find bins containing both the new Apple Audio Unit plugins and the older FCP audio plugin set.

Figure 10-14 Apply the 3 Band EQ from the FCP set to the audio portion of the sequence clip by dragging and dropping.

After applying the 3 Band EQ, double-click the clip so it loads into the Viewer. Switch to the Filters tab and click the widget to look at the EQ Filter. Under Audio Filters you should see the 3 Band EQ. Click the widget to look at the parameters (see Figure 10-15). ou will find a Frequency slider and Gain, or level, slider for each of the Low, Mid, and High frequency ranges. As described above, you set the specific frequency in each range and adjust the amount of boost or reduction applied to that frequency. If you don’t want to affect a frequency range at all, just leave the Gain slider at 0 dB.

How do you manage this filter and its parameters? What follows is a good workflow for isolating noise in a clip and then reducing it. Locate a short section of the clip in the sequence that contains dialogue, but also contains objectionable noise, such as hissing or low hums. Then, in the Toolbar, choose the Range Selection tool (keyboard shortcut, ggg). Go to the clip in the sequence, click inside the body of the clip, and drag. You will see that you are selecting only a portion of the clip (see Figure 10-16). Make sure that you’ve selected the area of the clip that you want to use for a test, about eight to ten seconds that contains the sound you want to keep (dialogue) and the sounds you want to eliminate (hiss or hum heard during pauses in dialogue).

After you’ve selected the area, go to Mark > Mark Selection (keyboard shortcut, Shift-a). This will put an In and Out point for the range of selected frames in the clip. Set FCP to Loop (View > Loop Playback, keyboard shortcut, Control-l). To play from the In point to the Out point, choose Mark > Play > In to Out (keyboard shortcut, Shift-). Thus, after setting the selection, go Shift-a, Control-l, and Shift- to begin looping a preview of the test section (see Figure 10-17).

One of the greatest improvements to FCP’s audio suite is noninterruptible filters. This means that as you make changes in the Filter tab, the playback in the Canvas updates and continues. You make a change in the audio filter settings, listen for the effect, then continue

Figure 10-15 The 3 Band EQ filter lets you boost or cut levels for three different frequency ranges.

to adjust until you arrive at the perfect settings. What sort of settings should you start with? Here are a few very common frequencies of annoying sounds that you will have to address at some point:

1. 8000 Hz Electronic hiss—the kind of hiss that doesn’t exist in the room, but is generated by the electronics of your microphone and the camera.
   

   Figure 10-16 Using the Range Selection tool (keyboard shortcut, ggg), click and drag in a clip in the sequence that contains both dialog and noise.

   

   Figure 10-17 After setting an In and Out point based on the selection within the clip and toggling loop playback, hit Shift- for Play In to Out. This will start the test section looping so that you can find the appropriate EQ settings.

2. 15,734 Hz Television refresh whine—if you record too close to a television monitor, you will very likely pick up this high pitched whine, too high to easily hear, but there and very annoying.
3. 60 Hz—a very low hum originating from problems in electrical wiring and grounding in the location you are plugging your equipment into. Once again, difficult to hear when recording, but painful when trying to get rid of in your tracks.
4. 20–200 Hz—low rumbles, often structural vibrations, and noise that you can’t ‘hear’ but which is recorded to tape and interferes with other frequencies.
5. 1000 Hz—a very common frequency for room hiss, and annoying sounds like heating and air conditioning vents. Unfortunately, this frequency is also right in the middle of the human vocal range, so be careful!

The Parametric EQ

The Parametric EQ works from a similar methodology. It isolates a frequency range, then lowers the level by the amount you tell it to in the controls. The difference here is that you only have one band in the Parametric EQ filter, rather than the three in the 3 Band EQ. To make up for this limitation, the Parametric EQ allows you to select the frequency band you are adjusting, plus it lets you make that frequency band larger or smaller. In other words, you can affect only one group of frequencies, but you can make that group of frequencies cover a range as large or small as you want to focus on eliminating the noise you hear.

Click the checkbox in the 3 Band EQ filter controls to check it off, and then go Effects > Audio Filters > Parametric EQ (see Figure 10-18). Take a look in the controls and you will see two sliders you would expect. The second slider control, however, is new. Q is the control that determines how wide the range of frequencies you control is. When it is set to 1, it is nearly as wide as the entire frequency range! When it is set to 20, it is covering a very narrow frequency range.

Figure 10-18 Disable the 3 Band EQ, then apply the Parametric EQ to the clip. In addition to the Frequency and Gain, there is a Q slider, which determines the width of the frequency range you are adjusting.

The smart way to work is to set it at around 3 (which is very wide), set the Gain(dB) to −20, and start moving the frequency control around as the clip plays until you hear the noise disappear. Once you find that range of frequencies, start adding to the Q value, and bumping the frequency control until you get Q up to about 10 and you are still hearing elatively little noise. Evaluate how the rest of your clip sounds with this EQ setting, and then make more changes if necessary.

If you need to cut down the high frequencies and the low frequencies simultaneously, without affecting the mid-ranges (which is, in fact, the most common audio clean up situation!), you would have to apply two separate applications of the Parametric EQ. Still, in many situations where there is only a specific noise source to clean up and which is spread across a range of frequencies that are too broad for the individual bands of the 3 Band, or even 31 Band EQs, Parametric EQ may be easier to work with.

The Voice Over Tool

Another really great audio tool is the Voice Over tool. This tool makes it a snap to add classy, high quality voice-over tracks to your video productions without having to record to tape, and then capture the material! The Voice Over tool is what we call a punch-in, or direct audio dub to your FCP sequence.

The Voice Over tool is very easy to master and incredibly easy to integrate into your workflow. Even those who don’t need voice overs for dramatic effect will find it useful. If two or more people are working on the same project, it’s a great way to leave messages and commentary about the way to deal with certain footage. Nothing could be more convenient than having a separate audio track that is especially reserved for notation. Need to tell someone else how to handle an edit? Bring up the Voice Over tool, select the target track, and talk away. It’s as easy as that.

Setup and Things to Consider

etup is pretty easy. The Voice Over tool can access nearly any audio input device that works with your Macintosh as long as it is compatible with Sound Manager, the QuickTime engine in the Mac OS that makes sound possible. Most USB Mike input devices qualify, such as Griffin Technologies ‘iMic,’ as does the old microphone input on older G3s and G4s. PowerBooks all have a built-in microphone mounted on the lid of the laptop; iMacs have a built-in mic as well. You can even hook up your DV camcorder through Firewire and use it. There are, of course, the expensive add-on PCI card solutions, which may or may not work, depending on compatibility with Sound Manager, but, in general, there are plenty of ways to get audio into the Voice Over tool.

To get access, go Tool > Voice Over (see Figure 10-19). When you do so, the Voice Over tool will appear in the Tool Bench. The first thing you need to check is whether or not your audio input device is available. Look to the lower half of the window for a drop-down

Figure 10-19 To access the Voice Over tool, go Tool > Voice Over, and then look in the lower half of the window to make sure the Source tab shows the device you want to use.

tab labeled Source. Depending on your Macintosh, and what you you have connected to it, you may see a number of options.

The likely choices will be Built-in Audio Controller, DV Audio, and USB Audio. The Built-in Audio Controller refers to your built-in 16-bit sound card that all Macintoshes have on their motherboard. Depending on the Mac you own, this Built-in Audio Controller may have a microphone input. If so, you can certainly use it for creating voice overs, though you will need to connect a microphone to the input. PowerBooks, as we said earlier, have such a microphone built into the lid of the laptop, and the iMac models also sport a built-in mike. The Input drop down below this will list the possible inputs that are available using the Source. If the Source drop down is set for Built-in Audio Controller, then the available Inputs will be Internal Microphone (for built-in mikes like the PowerBook) and Line In (if such an input is available, such as a built-in mike input).

DV Audio refers to a DV device potentially connected to your Firewire port. This will show up regardless of whether there is actually a DV camera connected or not. However, if there is no such device connected, when you select it in the Source drop-down menu, you will receive a message stating that it is missing. Don’t engage this option unless you, DV camera or deck is connected and turned on, not in VTR mode (as is normal), but in record mode so that the camera’s mikes are actually working. In addition, remember that you need to set View > External Video to None. Firewire cannot pass video or audio in both directions at once, so you have to disable Firewire output and allow the camera to send audio through the Voice Over tool.

Capturing voice over audio through your camera or deck is both better and worse than capturing through the Built-in Audio Controller. It is better than the Built-in Audio Controller because its sample rate can be set at 48K, higher than the 44.1K limit of the built-in sound card. In addition, the native 48K that is brought in from the camera will exactly match the sample rate of your sequence, unlike the 44. 1K clips brought in from the built-in sound card. On the other hand, it is worse, because on-camera mikes are some of the worst sounding mikes on the planet. They tend to pick up more noise than true mikes, and because they are attached to a bulky camera, cannot be easily positioned for the best possible sound recordings. In general, this will work in a pinch, but if you really care about the quality, you won’t use either of these first two Sources.

Rather, look into the large number of USB audio interfaces that have saturated the market in the last two years. USB audio capture devices are uniformly inexpensive, have easonably reliable, standardized drivers built for Mac OSX, and, generally, have quite good ecording capability. Most feature at least 48K audio sampling and the ability to get a lot further away from your loud, noisy, buzzing computer than the Built-in and Firewire nputs would. If you are going to do regular punch-ins, it’s worth looking into.

At the bottom of the Voice Over box, you will see a section labeled Headphones. It is strongly advised that you monitor your voice over delivery through headphones. As ou will see when you begin to record your voice over, FCP lets you hear the audio tracks already present on the sequence that your voice over will accompany when you are finished. In addition, it gives a countdown, a series of beeps that let you know when to start talking and when you should stop. If you let these sounds come through the Macintosh built-in speaker, they will bleed over into your recorded voice over track. You can adjust the level of audio feeding into your headphones here; it will not affect the recording at all. You can also disable the beeping cues, although this is a pretty bad idea, since you’ll be losing the ability to easily tell when you should start your delivery! There is a visual count down as well.

Setting the Offset

One important item you should initially setup when you start using a new audio input is the offset. Offset is a delay you can insert in your audio capture to compensate for latency, or the delay that occurs anytime you try and match the audio you are punching in with the tracks in the sequence you are laying voice overs underneath. This is going to be different for the various audio inputs available to you, so you need to check whenever you start using a new input. USB, DV Audio, and the internal sound card all have different delays, due to the different electronics they use to get audio into the Mac, so you need to run a quick test with whichever input you are using.

On an empty sequence, navigate to the tenth second by typing in the number “1000” and hitting enter (see Figure 10-20). Then hit the I key to create an In point. Type the number “2000” and hit enter to navigate to the twentieth second. There, hit the O key to enter an Out point. Now, hold the recording microphone up to the headphone speaker and hit the Record button in the top left corner of the window. You will be letting the Voice Over tool record its own cue beeps.

Continue holding the microphone to the headphone speaker until the voice over recording has completed. After a moment, the newly recorded track will appear on the sequence. Use the Command-Option-W keyboard shortcut to enable waveform previews on the sequence (unless you already had them enabled). Use the Up and Down Arrow keys to navigate to the end of the voice over clip. Use the Command − + keyboard shortcut to zoom in on the sequence until you are seeing increments that are individual frames.

Make sure that your playhead is on the frame ‘00:00:20:00’ then count the number of frames after that frame number that the last beep’s waveform ends (see Figure 10-21). The number of frames difference here is the correct number to use in the Offset drop-down tab. After you set this, do another quick test run to make sure that this was the correct offset. After you get it set, you can be sure that your delivery will be right in sync with the video you are doing a voice over for. Apple states that USB audio input devices tend to have an offset of one frame while DV cameras and other Firewire input devices may have offsets of up to three frames.

The Rest of the Voice Over Window

Furthest left is a practice run through button. When you hit it, the Voice Over tool will run through the exact range of frames included in the In to Out point. If you have not set an In to Out point range, it will use the current position of the playhead as the In point and the last frame of the sequence as the Out point. Clearly, it’s a good idea to limit your range of frames to just what you want to cover, so set In and Out points.

We have already used the red Record button to test the Offset of our mic and input device. Next to it, on the right, is the Discard button. This button is only available just after you have completed a take or voice over recording. If you want to immediately toss the last attempt, you can hit this button. A warning will come up telling you that this is an irreversible act. And it is. The take will be wiped from the disk, so make sure you don’t want to keep it. To the right of this is the progress meter. In this box, you will receive messages telling you how much longer you have, as well as a percentage meter informing you how much time is left during the voice over take.

Underneath this is the Audio File section. This box informs you of the currently set target track where the voice over clip will be sent. The Name text field underneath the Target label is for assigning a name that will be applied to the voice over clip to be created in the sequence. If you don’t enter a name, the Voice Over tool will use the last name that was assigned plus a number.

Figure 10-20 To set the offset, create an empty sequence and navigate to the 10th second (00:00:10:00) and set an In point. Set an Out point at the 20th second (00:00:20:00). Then, make sure the mic is near the speaker and hit record to punch in a test beep track.

Figure 10-21 The number of frames discrepancy between the Out point and the beep’s spike in the audio waveform is the required offset (you might have to trim the end of the punch-in clip to see the beep in the waveform).

Finally, in the Input section, in addition to the Source and Input tabs, there is a level meter so that you can detect any clipping in the recording you are making. If you see clipping, first make sure you are not speaking too close to the mic! Many mics are very sensitive, and you need to add a little distance in order to get a good recording. If the levels are still too high, you might drop the Gain control somewhat and see if the results are more successful. Finally, look at the Rate tab and make sure you have selected the highest sample rate available, since some devices can use several different ones. If you have options, 48K is the preferred sample rate, since this will match the rate we are using in the sequence.

Using the Voice Over Tool

To put the Voice Over tool to work is very simple. First, you need to set the range of frames ou want to record the voice over for. Find the frame where you want to start your delivery on the sequence and hit the I key to insert an In point. Navigate up to the frame where you want the voice over to end and set an Out point. Unless the delivery must be in one unbroken piece, consider breaking the range up and doing several individual voice overs.

Once you have set the In and Out points, take a moment to figure out where the voice over clip should be sent. It is not critical to specify this. FCP will never record a voice over into a track where there is already a clip present. If a clip is in the way, FCP will create a new track for the voice over and bump the tracks already containing clips down into new tracks. No audio sync will be threatened. But it’s a good idea to organize and plan ahead. In general, for organizational purposes, you will want to dedicate a couple of audio tracks—out of the way of your normal audio tracks—that are linked to your video tracks.

The rule is that FCP will always use the ‘a2’ source track. This is not Audio Track 2 necessarily; this refers to the Source/Destination track router. If, for instance, you have set a2’ source track to A5 destination track, the Voice Over tool will select A5 as the target track for the voice over clip. If there is currently a clip in the target track, the Voice Over tool will move the ‘a2’ source track to the next track down. If there is also a clip in that track, the Voice Over tool will leave the target set for the track, but will bump the other tracks down a step out of the way.

To begin the voice over recording, put on your headphones and hit the red Record button. You will immediately hear a countdown start beeping. It will stop a few seconds before the frame you wanted to start your delivery on. In the background, FCP starts ecording a few seconds before you indicated you wanted it to in case you begin a little early, so that you don’t accidentally cut off the beginning of your first word. It also continues for two seconds after the last beep to make sure you aren’t cut off.

The Canvas window will play through the frames of video that your voice over is to accompany. In the Progress meter, you will see the countdown to the end of the voice over. If the voice over range is more than fifteen seconds long, at fifteen seconds before the end of the voice over, you will be given a warning beep. Then, in the last five seconds of the voice over range, you will hear five warning beeps, followed by a ‘finished’ beep. If these beeps

Figure 10-22 After the voice over take, the clip appears in the correct target track.

get in the way of a decent voice over delivery, feel free to disable the “sound cues” checkbox in the headphone. Just be careful to monitor the Progress meter so you know where you are in the recording session.

After the recording is finished, the clip will appear in the track (see Figure 10-22). You can give the track a quick listen, immediately discard it, or temporarily mute it using the keyboard shortcut Control-B so that you can have another go at the voice over without hearing the previous take. Immediately discarding the take can be a bad idea; usually you will not get an entirely perfect take in one go. If you complete several takes, you may be able to piece the best parts of each take into the perfect voice over; you’d be surprised to hear that this is frequently the way that professional voice over tracks are cobbled together!

After you get the voice over completed, you can eliminate the takes you don’t want to use, and trim the ones you do want to use. You can move the good takes up into more accessible audio tracks. As a final note, you might want to be careful to back up these voice over tracks. Unlike your video and audio tracks from your camera, these voice over tracks have no timecode. If anything happens to your hard drives, you will have lost them forever. Use the export techniques in the next chapter to export the voice over audio, then burn them to CD-ROM or other back up media for safekeeping.