help with correct audio levels

[John Fishback]

The flattened tops and bottoms indicate the sound is clipped. As a result, it will sound distorted. This happened at one of two times, either during your original shoot as the audio was recorded to your camcorder or when you transferred the sound from the camcorder to your edit app. Hopefully, it was the latter as that can be corrected. However, if the original sound you hear out of the camcorder is hot/distorted, then you’re stuck.

It wasn’t clear from your description if you tried to recapture the audio or just lowered the clip level. Lowering the clip level in the edit app won’t do anything if the audio is already clipped. Try capturing again and set the audio to a lower level than before, and see if this helps. Also, how are you capturing the audio? Are you using a mixer? If so, make sure the mixer’s input is set properly so it’s not overloaded.

Your meters or level indicators don’t always tell you if everything is allright. Any recording device has a series of amplifiers that handle the sound. The gain control usually affects one of the last stages of amplification. So, if the sound being fed to your recorder is too hot (loud) it may overload (clip) the first amplifier (preamp) in the series. Even turning down the gain in a later amplification stage won’t change the fact the intial audio amp is overloaded and distorted. The old Sony BVW-35 field Betacam recorder was famous for its audio being easily overloaded by feeding it too loud a signal. Any time I saw the audio gain controls set below 5, I questioned the sound person to make sure the input level wasn’t too hot.

Bottom line: Always listen to a playback of your audio after the first take (or do tests before the first take) to make sure it’s okay. (It’s not a bad idea for video confirmation, too.)

John

Dual 2.5 G5 4 gigs RAM OS 10.4.3 QT7.0.3
Dual Cinema 23 Radeon 9800
FCP Studio 5 (FCP5.0.4, DVDSP4.0.2, Comp2.0.1, STP1.0.2)
Huge U-320R 1TB Raid 3 firmware ENG15.BIN
ATTO UL4D driver 3.50
AJA IO driver 2.1 firmware v23-28
SonicStudio HD DAW, Yamaha DM1000, Genelec Monitors

[Ruby Gold]

Thanks for the response John. I’m capturing directly from the tape (in my clamshell) into the computer and editing in PPro 1.5. Sadly, the capture settings in PPro don’t include the ability to adjust audio levels.

The thing I don’t understand is if the audio is clipped, as indicated from the flattened waveform, why it’s not registering as so in the audio mixer in PPro. It doesn’t ever go up into the “red” clipping area on the mixer when played back in the timeline–yet much of the waveform is flattened.

Thoughts?
thanks-
Ruby

[John Fishback]

It was probably clipped when it was recorded during the shoot. Once clipped, it will stay that way regardless of subsequent level settings. One practice when shooting is to send the same audio to channels 1 & 2, but set Ch 2’s level down 3 to 5 db. This way, if Ch 1 clips, Ch 2 might be okay.

John

Dual 2.5 G5 4 gigs RAM OS 10.4.3 QT7.0.3
Dual Cinema 23 Radeon 9800
FCP Studio 5 (FCP5.0.4, DVDSP4.0.2, Comp2.0.1, STP1.0.2)
Huge U-320R 1TB Raid 3 firmware ENG15.BIN
ATTO UL4D driver 3.50
AJA IO driver 2.1 firmware v23-28
SonicStudio HD DAW, Yamaha DM1000, Genelec Monitors

[Ty Ford]

Couple of thoughts.

Are you bringing the audio from the camera into your laptop as analog audio? If so, there might be an adjustment you can make. Obviously the audio is clipping somewhere.

“”Some audio workflows must be calibrated at a specific tone level. The default level of the 1-kHz tone is 012 dB referenced to 0 dBfs. You can customize the tone level to match your audio workflow by choosing Clip > Audio Options > Audio Gain with a clip selected.””

I’m guessing you mean -12dB not 012dB. 012dB doesn’t make sense. Have you done as it said and selected a clip and “audio gain?” That may calibrate the system. Do you have access to a system that will import the audio and video without going through the analog inputs? If so, try that.

Or, hope not, the person mixing was looking at RMS (average) leveles and not Peak levels on her mixer when she was recording. An experienced sound person would know the difference, but ya never know.

-20 dB, BTW, IS a good place to set the 0 VU tone from your mixer, especially if you expect outbursts. Make sure you meter is set to PEAK, not RMS. If it’s set to RMS, it will look low and you’ll want to jack it up. If you DO, without checking for PEAK level, you can very easily clip the audio.

I use a mixer with a very good limiter. The mixer reads RMS and PEAK simultaneously, so I alwsys know what’s going on. If I’m feeding an analog camera, I set my limiter to a +4 or +6 threshold and send to to 0 VU. The audio to the camera nevert goes over +6 dB, which most analog camera can handle.

If I’m feeding a digital camera, sometimes I’ll leave the limiter set at +6 dB and send my 0 VU tone to the camera. Then I’ll set the tone at the camera to -12. Because my limiter will catch anything above +6 dB, except hand claps and very quick sounds, I’m fine. In t that situation, I have 12 dB above 0 on the camera for peaks and my limiter will catch anything or +6dB, leaving me 6 dB between 6 and 12. If I hit the limiter a little, that’s OK.

Regards,

Ty Ford

Regards,

Ty Ford

Ty Ford’s “Audio Bootcamp Field Guide” was written for video people who want better audio. Find out more at https://www.tyford.com

[Peter Groom]

HI there
I think the comprehensive comments from Ty and John just about cover this one, but for my addition.

1) If it is recorded, too hot, or clipped as a digital signal then all of the bits have become full. ie the digital datastream that is the audio instead of being 0’s and 1’s becomes all 1’s. This is where clipping occurrs and is one of the biggest negatives of the digital domain as the audible effect of clipping is far more objectionable to the ear than analogue tape saturation onset where the effect is gradual.
Like a ming vase dropped down the stairs, once its broken , it is broken and no amount of tape and glue will mend that audio.

2) Re recording levels. The recordist ought to have laid down some tones from her mixer at the head. These were stated to be -20 Correctly. That represents a 1k tone = PPM 4. The audio then following ought to follow at @ +8DB so peaking at -12. There is still a good way before the audio hits clipping at the recorder stage. If the waveform displays show a flattening of peaks but there is no audible clipping, and the audio was passed in digitally, then Id suspect that a limiter either in the remote mixer, or some sort of on board camera AGC was responsible for the problem.

In general, if theres a recordist with an external mixer and tones etc are employed then there should be no need for AGC at camera stage.

Peter
Dubbing Mixer