Technical Aspects of Shooting to Sync Playback:
Non lip sync sequences for Music Videos are relatively easy. Since the images do not require frame to frame alignment with the music and lyrics, the filmmaker is pretty much free to do whatever they want. Footage can be shot at any frame rate (overcranked for slow motion, undercranked for accelerated motion, stop-frame, etc.) Sync playback of the music during shooting is not a requirement, although many filmmakers prefer to listen to some sort of rough playback track (also referred to as a tempo track) to guide them in terms of shot length (for eventual cutting to the beat) and pacing of camera moves (to match the mood or pace of the music).
However, when we are shooting a Performance sequence for a music video, then strict adherence to sync playback rules must apply.
When we record sync dialog for a movie, it is critical that the recorded audio be able to match the played back (video or projected) picture frame for frame, or else the actors will appear to be out of sync. Being off by even a fraction of one percent is enough to ruin a scene. That is why professional motion picture cameras run at speeds carefully regulated by crystal controlled motors and the production audio recorders rely on complex sync and/or timecode systems to completely eliminate any speed variance (wow & flutter). Both the film camera and the audio recorder capture the scene in REAL TIME so that they can be aligned later during editing and played back with accurate lip sync.
The traditional clapstick slate (or modern timecode) is used to line up the respective audio track with its matching picture. Of course, if audio is originally recorded simultaneously on the videotape, as in a video camcorder, then lip sync is guaranteed since we do not need to concern ourselves with matching up the track from an audio recorder with that from the camcorder.
When we shoot a music video, the audio is NOT recorded live on the set, but is instead PLAYED BACK for the musicians to pantomime in lip sync to. It is sort of like recording live dialog, but in reverse. The audio track must be able to match the camera track frame for frame, or else the lips will be out of sync. This requires a camera with crystal controlled speed, and an audio playback system completely devoid of any wow & flutter. If the audio does not play back at precise speed every single take, then the musicians will be pantomiming to a song that may be several beats off tempo from the actual recording (the actual song that appears in the finished music video).
To do sync playback, you will need a recorder or playback device that is capable of speed accurate playback every time. That means either a professional grade CD player (not your $29 pocket unit), a digital recorder (DAT, memory card, hard-drive), or a Nagra analog sync recorder. In addition, we need some way to record a signal onto the camera to allow us to know what section of the music was playing while we were filming.
Here is how we do it in the professional world…
The first step is to obtain a copy of the final music selection that the artists will be performing to in the music video and create an EDIT MASTER. That means making a digital copy of the music that has SMPTE timecode added to it. The timecode should run at 29.97 NonDropFrame rate. This piece of music will become the master soundtrack that we will edit our music video images to.
Multiple copies of the EDIT MASTER (with matching timecode) will be produced for us to playback in the field during production. Format will match the type of audio device that we will be using for sync playback. These copies are called PLAYBACK DUPES.
In the event that we are using a professional CD player, the PLAYBACK DUPE will be mixed down to mono, so that the CD will have music on the left track and SMPTE timecode on the right.
Out on the set, the Director will call for the camera to roll first, and then call for the Playback. As the music begins playing, we need some means to identify the musical section for the camera. This is professionally done by means of the timecode that was pre-recorded alongside the music. A timecode slate is used to display the running timecode for the camera to photograph. This provides a match frame for the video editor to use in order to align picture with the appropriate point in the master song.
After shooting a brief amount of the timecode slate, the Director is free to call for Action.
A master angle is usually shot, covering the entire song. After that, only selected portions of the song will be played back in order to get good close-ups and inserts. Only a brief lead-in of the music will play to allow the musicians to get into the rhythm, continued by the section that the Director needs to film.
Although not recommended by professionals, it is possible to shoot a music video in VIDEO without timecode, since audio can be recorded in the camera. In this case, the camera would hear (either through a mic or a direct connection to the playback device) the section of music being played back. Later, the editor would just have to manually line up the audio on the video against the master soundtrack. It is not as fast as doing it with matching timecode numbers, but it can be done. Note that the audio recorded on the camera is deleted from the edit as soon as the picture is aligned with the master soundtrack. Never try to make a music video by patching together audio snippets of the song, even if they were recorded clean onto the camera.
Shooting a music video is a long and tedious process. It only takes one slip by one musician to screw up the lip sync, play to the exact note sync, of a take.
A short, two or three minute song can easily take days to shoot!
Audio Playback Techniques:
The musicians need to hear the music being played back. Usually, this is done by outputting the playback device to a conventional amplifier & loudspeaker system. Professional musicians are often partially deaf from all their exposure to amplified music, and may demand that you use concert loud volume for playbacks. If so, make sure that you wear hearing protection (insertable earplugs) so that you and the rest of the crew don’t end up like them! Most of the time, though, you ought to be able to get by with just audible playback levels from a small speaker/amp system.
Sometimes it will be necessary to record live dialog while a sync playback is going on in the background (more often done in feature films than music videos, but some artists do want to slip in a line of dialog or two). If this is the case, then you must be able to provide silent playback.
Silent playback may be as simple as just having the musicians wear small earpieces. Wires go down their necks, under their wardrobe, and eventually plug in to a headphone distribution box.
If the musicians (or dancers) are scantily clad, then we can use induction earpieces. These are similar to wireless hearing aids; they pick up audio transmitted by a local antenna. In the example of hearing aids, these wire loop antennas are worn around the collar region underneath clothing. They broadcast a weak signal up to the earpiece; hence there are no visible wires.
For music videos, we run an antenna wire completely around the set and power it from a 200 to 400 watt audio amplifier (usually used to drive loudspeakers). The antenna wire is simply regular hobby wire connected to the speaker terminals of the amp. A resistor or two is added to the wire so that the amp does not burn itself out looking for the loudspeaker. Anyone located within the perimeter of the wire antenna will be able to hear the music (or instructions) in their miniature, hidden-in-the-ear earpieces.
Another silent playback technique is to replace the loudspeakers with sub-woofers. These sub-woofers are adjusted so that they can only put out extremely low frequency sounds, usually under 60 hertz. The musicians and dancers can feel the beat, but the pulses of the music are too low in frequency for motion picture microphones to hear.