Nov 15 2009
If you’re reading this article you probably are aware that shooting HD video with DSLRs is all the rage among digital filmmakers. There are far more reasons to shoot this way than not, but one of the problems with DSLRs are their sound recording capabilities. THe Canon 5D Mk2, the Canon 7D and the Panasonic GH1 have external microphone inputs but unfortunately these use auto gain control and have no manual control over input levels. Also these cameras don’t have headphone monitoring.
The most common solution to these problems is to use film style second system or double system sound.The advantage of second system sound is that you have a redundant high quality sound recording that you can sync to your camera images in post-production. The methodology for second system sound ranges from very simple to incredibly complex methods. [See this article which outlines no less than 16 different methods for doing second system sound. ]
Here is my simple method and workflow for capturing and syncing great sound for HDSLRs.
First you will need some sort of external high quality sound recorder. It is important to note that your sound recorder doesn’t necessarily have to be a sound recorder. I have several DV and HDV camcorders that have microphone inputs and headphone monitoring. So one possible method would be to use one of these as a 16 bit 48KHZ sound recorder and as an added bonus you may even get some usable B-rool video out of the deal. The other method is to use a dedicated sound recorder, one of the coolest and most popular of these is the Zoom H4n recorder. The H4n is a high quality compact rugged and versatile machine and at $300 it’s a bargain. If you don’t have one or something similar, you should buy one. Of course, you may have a portable DAT machine or some other legacy sound recorder, and any of these will work.
Secondly I’m going to presume that you can hook up a high quality external mic to your sound recording device, whether it is a shotgun mic on a boom or wireless lavalier. The ideal situation would be to have a dedicated sound recordist/ boom operator on your shoot, but if you’re a one person operation you can still use a camera mounted mic hooked up to your sound recorder.
Okay, this is the most important part of this method: DO NOT DISABLE SOUND RECORDING ON YOUR DSLR! You are going to use the sound as your reference audio. I edited a cooking show where it was a two camera shoot and the Steadicam was the A camera. Unfortunately the Steadicam didn’t record sound and they forgot to slate about half the shots. If I had reference audio for that camera, I would have saved the producer about 20 hours on that edit. So the lesson is, ALWAYS RECORD SOUND. I have a Audio Technika ATR55 shotgun mic hooked up to the hot shoe of my Panasonic GH1 and have found that actually 60% of the time, I’m using the audio recorded in camera. I”m still recording second system sound, and even when I use the camera audio, it’s nice to know that I’ve got back up audio if I need it. Also when you factor in how many of your shots are MOS or don’t need sound, you’ll be surprised how little post production syncing you actually have to do. The only thing to remember, and this is very important, is that if you can’t slate your shots using a clapper, is to be sure to have your talent clap or snap their fingers or snap your fingers when you are rolling before you shout action. ( You could sync to “action” but believe me, a clap is better.)
So lets get to the simple method of syncing the sound. I use Final Cut Pro, so I’ll outline the method using FCP, but other editors can use much the same method.
Import your video and convert it to whatever frame rate you are going to edit in. Since we are not using time code to sync, frame rate doesn’t matter. Let me repeat that; frame rate doesn’t matter.
Next, you want to import your audio in to your video editor. Hopefully your audio is at least 16bit 44.1KHz PCM audio. Recording as MP3 is not acceptable. If you are using a Zoom H4N you may even have 24 bit 96KHz audio, which is overkill in my opinion. I think that 16 bit 48Khz is the best compromise for sound quality, file size and render speed.
Once you have both your video with reference audio and your second system audio imported, go ahead and drag a shot to the timeline and be sure to include the clap or snap on the reference audio track. Next drag your second system sound onto the timeline below your reference audio track. Then line up the clap or snap by ear or even the waveforms on the timeline ( make sure you have show waveforms enabled). You can playback your tracks and if you hear flanging, nudge the second system sound track (by using the option left or option right arrow keys in FCP) until the sound is in sync.
Once you sound is in sync with the reference track, trim your second system sound track to the same length as the reference track, then select both the video track and the second system sound track and use “Command L” to link it to the video track in FCP. Now that you have your second system (and hopefully better quality) sound track synced and linked to your video shot you can trim and edit the video at will and the second system sound will stay in sync. Your reference audio track is still linked to your video and while in most cases you will mute that track in the final edit, it’s nice to know that you still have it there and in sync if you need it. Lastly you will have to repeat the sync process for all of your master shots, but once you’ve got it down, it’s relatively painless and you don’t have to do any math.
I will note here that Beachtek Audio makes a great product for single system sound recording on DSLRs called the DXA-5D. in addition to having high quality XLR mic inputs, it also tricks the auto gain control in the DSLR using an ultrasonic tone. The only trouble is, they are out of stock at the time of the writing of this article.