I hate to think that I had been procrastinating editing some videos I shot for some friends, because I really had not been, although the work required a task that really bugs me. Despite my DSLR cameras being great for recording HD video, their audio capabilities are well, open for improvement. So I do what lots of videographers do. I let the camera mic run and also use external mics with an outboard audio recorder. My rig is very simple. Basically I use Sennheiser wireless lavaliers into a Zoom H4N field recorder and save the files as WAV format. This gives me very high quality audio that I can lay into my non-linear editor as an extra track beside the video with embedded audio from the DSLR. The Canon DSLRs also record audio through either the built-in mic or a frame mounted shotgun. I had tried the Beachtek box and while it works, I've had issues, and even though the Rode Videomic is very good, it's one mic and doesn't give nice stereo separation when shooting a two person standup.
I am using Final Cut Pro X which was recently updated by Apple to 10.0.3. This brought support for XML controls and that enabled the use of a tool I had looked at once in the days when I used Final Cut Studio. The tool is called PluralEyes and comes from Singular Software. You can download a trial of the product here. If you want a non-integrated audio track tool, they also do a tool called DualEyes.
FCP X purports to be able to sync multiple audio tracks together. It can. Mostly. With multiple retakes and inconsistent start times between audio and video, I have found that it works ok, but not perfect and not necessarily consistent. PluralEyes is a special purpose tool, that carries a price tag of $149 USD. So it's not like the free that's built in to FCP X. You should expect something really great for that kind of money, and my experience today says you get it.
I had eight separate videos to edit, each with multiple takes and audio files. So long as I kept them straight, PluralEyes did the job quickly and effectively. While it is a standalone app, it integrates very nicely into your FCP X workflow.
Here's the workflow I used today.
- Create a project inside an event in FCP X for each video.
- Find all the segments and import them into the Event Library (what we used to be call the bin)
- Open each audio segment in The Levelator and let it do it's magic to even out the audio levels. I use the Levelator all the time for editing spoken audio content. It's awesome and free.
- Place the video and audio clips in the timeline for the project and line them up appropriately
- Step up a level to the Project Library and do a File | Export of the XML for the project
- Launch PluralEyes and load the XML file created above into it.
- Choose SYNC. I prefer not to have it replace the camera audio even though this creates a step for me.
- Once done, and it is very quick, a new project is created with the same name as the original with the word synced appended. Go into this project and you'll see your video and audio tracks properly aligned and in sync
- Select the combination track from the camera and Detach the Audio
- Delete the camera audio track
- If your audio is only one sided stereo, change it to dual mono to load both Left and Right channels. Adjust overall levels if needed using the db drag line in FCP X
- Select the camera video track and the good quality audio track. Now make them into a Compound Track. This makes editing much easier.
- Cut the compound track to your liking, adding your transitions, intros, outros and the like and finalize the project.
That's it. It's like working with a professional video / audio track from pro gear. PluralEyes saves a lot of time and makes life much easier. I have not discovered a down side, other than the license cost of the software, but for the hours it saved me, it's worth it. You can try the software for free before you buy which is very good of the manufacturer.
PluralEyes is available to work with Final Cut Studio, Final Cut Pro X, Avid Media Composer, Adobe Premiere Pro, Sony VEGAS and Edius. The 30 Day trial is not impaired in any way that I discovered.