Post-production improvements of classical concert
 

Greetings...

First off, I'm new to filming and editing, so I don't have pro gear to work with, just a Canon HF S100.

Back in December, I recorded a choral concert as a favor for the group. I was at the back of the concert hall (on the balcony) using the built-in microphone. I'm editing in Adobe Premiere planning to output H.264 for YouTube.. In my first attempts, the audio sounds a bit muddy. What are some ways I could improve the audio in post? I figure I can add EQ to the audio, but I'm not sure which frequencies I should be messing with. Thoughts?

Here's the video - YouTube - Christians Be Joyful - From the Christmas Oratorio by J.S. Bach

Any additional hints/tips would be much appreciated!

-Adam

Is that a mic stand with a stereo pair on it, right in front of the stage? If so,I would suggest that you get a copy of the recording that was made if that is indeed what was happening there. Your audio is a lot better than I thought it was going to be (the quality of the performance is a big asset) but there's still not a lot you can do to improve it.

You won't want to hear this, but an on-camera mic at the back of the hall is the last thing you need for a recording like this. If there is another recording available, done from where the mic stand is, it's probably a lot better than yours and you should be able to sync it to the video.

Sorry I can't be more helpful.

I agree with Colin.

That is definitely a mic. stand with an ORTF pair by the looks of it. So find the person who did that recording - get a clone of the file and sync. it to the video.

Again, like Colin, I was surprised that your audio was as good as it was, given that it was a camera mic. at the back of the hall. But there is not much you can do with it.

When you are recording a concert, treat it like an audio recording and get the mics right and then feed those back to the camera - it's not worth doing any other way. The sound is far far more important than the picture for such things and this should be paramount.

So - I really hope you can find the chap who did the audio and get a copy for the video. It does look very official as the cable from the mics goes straight out and under the stage and probably out the back to someone doing it properly (rather than someone in the audience with a portable). The mics do seem a little low for best results, but they are at about the right distance, I think, and the recording should be very good and perfect for the video.

Oh - I just noticed that you did some zooming and panning. This is a definite no-no if you use an on-camera stereo mic. as it messes up the sound and is nasty to listen to as the sound wobbles about as you move the camera.

Thanks guys.. I was already thinking about finding out who has the recording, and seeing about getting a copy.. Syncing should be interesting.

Also - I agree about the zooming and panning, thought that myself right after looking at the footage. It would make sense if I had two cameras, but I'm upset with myself now that I did it. That's how we learn, right?

Thanks again.

Actually, synching should be a snap (pun intended). Just throw both tracks on your timeline and match up the waveforms. Even if they drift a little in relation to each other, you should be able to stretch to compensate.

Hi Adam.

When I first began to play this video I was under the impression that you were taking a feed from what appears to be a crossed pair at about a descent position. As to whether its an ORTF Schoeps I don't have that clarity of sight. If it is, then whoever is doing this audio recording is just about spot on.

But it always comes down to the performance space acoustics. Better acoustics and I would expect to see that pair further back to capture the nuance of phase (seperation) and natural delay.

The recording I hear has no seperation...

What I am hearing is pretty bad nasties at the LF end (the Tymps, due to room modes - natural resonences caused by parallel reflecting surfaces) and some phasing artifacts at the mid to hf end with the choral.

You can't Eq out this issue.

If, as you say you were recording the audio from the camera then you are hearing what the camera hears - at that position. And any slight movements with the camera (relating to audio HF reflections) will translate as phase shifts - plus other reflecting phase issues.

But you are also posting to YouTube. They are constantly turning the knob to save bandwidth, processing times and what else? I don't know because only YouTube know.

Problem for evaluation is not being able to hear the original, only that presented via a band-limited internet channel.

If this recording was gained from your camera mic, then I think you have done reasonably well. All that remains if you wish to improve matters for the future is to speak to a sound recording professional (or very knowledgable amateur) and co-ordinate the production.

Perhaps considering a sync recording technique with seperate audio recording might improve matters the next time. Then all you have to consider is a syncing issue, not one of audio quality.

:)

OT but belated welcome Claire. There's a whole bunch of dvinfo Brits on this thread (this will take you near the end) http://www.dvinfo.net/forum/these-pe...ml#post1488252 Apologies if you've found this already.

Claire, thanks for listening in so very closely. I am clearly not the educated audiophile that you are, as I couldn't pick any of those details out!

True that I can't expect a whole lot from the built-in mic on my camera, but I'm glad to hear it's not *that* bad. I'm still trying to get access to the original audio recording, as that would be my best option.

It is too bad that EQ wouldn't help much..

Don't know that any of this helps but...
 

I can't identify the specific microphones, but in listening to that again with the HD option on the video, it looks more like a crossed pair, vertically offset (not that it matters). I would opine that the relatively good quality of the audio under less than ideal conditions is a tribute to the auto gain circuitry of the camera which can just sometimes do a better job than an inexperienced operator using manual gain.

Another factor is that the ensemble is particularly well acoustically balanced - well done the conductor and the (probably scratch) orchestra who are particularly well in tune (bravissimo, D trumpets)! Just a pity we can't hear the organ continuo. The occasionally flat sopranos could have been corrected in a studio recording but it's not really that noticeable. The (as usually considerably outnumbered) tenors and basses are doing a sterling job.

(Ignore all this by all means - I'm being a bore. Almost certainly too much unnecessary information, and some of it opinion rather than fact).

:-)

Attached pic of mic array:

Bravo!
 

I have little to add that hasn't been mentioned already. I specialize in this kind of chorus and orchestra recording and/or video work. As others have mentioned, the audio wasn't as bad as some I have heard under similar circumstances.

The Bach Christmas Oratorio is one of my favorite Christmas works (and "Jauchzet, frohlocket" my favorite chorus from it.) The performance was quite good for (presumably) an amateur chorus and orchestra. Indeed, including some zooms and pans would have made the video record a bit more interesting. It helps to know the music so that you aren't zoomed in on a trumpet when they aren't playing, etc. :-)

Of course if you had a proper sound crew and equipment and half a dozen cameras, you could do something like this: YouTube - Thomanerchor: "Jauchzet frohlocket" by Johann Sebastian Bach :-)

The four things that work against you with your technique include...

1) DISTANCE - your location in the balcony is typically too far away to get a good stereo recording of that kind of source in that kind of room. Microphones can't use the kind of ear/brain processing that we enjoy unconsciously. The mic position must be artificially "tuned" to achieve the kind of balance that your brain produces for you automatically. There was a bit of stereo separation if you listen on good headphones.

2) LOCATION - We typically try to mic the chorus and orchestra separately in order to achieve some (perhaps minimal) amount of balance between the forces. The stereo pair on the visible stand may have created a recording that favors the orchestra at the expense of the chorus. I would have tried to get the mics higher and closer, but the producers of programs are not always cooperative with getting good recordings. :-(

3) MICROPHONES - The microphones built into camcorders are not noted for stellar quality. Most of them are there for marketing purposes, just to say that the camera has built-in microphones. They are typically 69-cent electret capsules. Furthermore, they are usually way too close together to produce much of a useful stereo effect. But of course, you can't expect otherwise in a compact form-factor of a small camcorder.

4) COMPRESSION - Most camcorders use "Auto-Level" to prevent clipping the audio track. On some camcorders this can be switched off. Auto-level is NEVER appropriate for recording music of any kind. It destroys the dynamic range (both short-term by "pumping", and long-term by reducing the contrast between ff and pp)


I agree that there isn't really anything you can do in post-production for the audio track. Certainly I would encourage you to contact and work with whomever was recording with the central stereo mic pair. If you can acquire that sound recording, you may be pleasantly surprised at how easy it is to sync. You would want to lay the audio track into the NLE as the master reference and "pull up" the video to sync with the audio. Trying to do any kind of "stretching", etc on an audio track is very difficult without producing undesirable audio artifacts. It is easier to "stretch" video because most viewers won't notice occasional single-frame repeats or drops.

But this is a great first effort. Would that my first similar recording were that good. Kudos and thanks for sharing your video. I enjoyed it.

Damned if I can tell much about the mics from watching - other than noticing that there seemed to be a couple of mic stands on stage as well. One looked like it had a dynamic mic on it for announcements etc but couldn't clearly see what was on the others.

Location of the large mic stand did seem to be pretty good, but as has been said, a bit too low. In a situation like this when the group is elevated and I'm on the floor I sometimes put the mics up on a 12 - 15 foot light stand just to get well above the conductor's head. Anyhow, without hearing the other recording it's hard to know for sure but I suspect it was pretty good - of course, the type of mic used would play a part in the result.

Re favoring the orchestra over the chorus - I'm not so sure as it isn't in total such a huge group and I think part of the conductor's job is to make sure the balance is right at his position.

And like everyone else has said, I think the sound is surprisingly good considering how you recorded it. I've sometimes actually had good luck being about 20 - 30 feet back in the audience depending on the characteristics of the hall, although you do get a lot more audience noise, coughs, program rustling, fidgeting, etc.