Any advice on my instructional videos?

[DavidWL]

Hello,

We're filming tango instructional videos, and are getting ready to do our 2nd batch of videos. Sample video: http://learntodancetango.com/video/35863752/#section=comment

Right now we are using 1 wireless mic (VHF audio technica), and 2 video camera's (Panasonic HDC-TM700K). We are going to probably add 1 camera to get feet shots, as well as moving to 2 wireless mic's, so she is miced as well.

Any advice? Particularly on how to improve the sound or the overall video quality.

Thanks!
David

 

[Alex H.]

Audio from the wireless is overmodulated (distorted)... and it's analog distortion, not digital clipping, which means that the problem lies somewhere between the lavaliere mic and the camera's input. You need to work on gain staging, or aligning each piece of the signal chain to send the cleanest and strongest signal possible to the next piece.

The first thing is that you have headphones connected to the camera at all times... especially when recording. If you aren't listening as you record, there's no way for you to know what you've got until it's too late to fix it.

Start with the transmitter, if there's a level adjustment there (even better if there's a clip indicator, usually a red LED). Since you didn't give the model number of the wireless system, I'm just guessing here. Have the talent speak at "performance level" (in other words, not mumbling or whispering). Bring the input level up just until you see clipping on the peaks, then dial it back just a tad.

Is there any kind of level meter or clip indicator on the receiver? On these cheaper systems, not usually... just curious.

With the receiver connected to the camera, and listening through headphones, bring the camera's input level up until the audio meters hit the reference mark (not sure what that is on the TM700, but it's usually some sort of noticeable mark on the meter, or a change in color of the meter segments). This reference mark is where you want your audio to average. It can peak a little above, and drop below, but it needs to spend most of its time hanging around that point.

Before you buy a second wireless system, you might want to read the "$500 Wireless Question" sticky at the top of the Audio forum. Just sayin'... there are better things out there than the cheap AT VHF systems, and you get what you pay for.

As for the feet of the dancers, a shotgun at floor-level, just off-camera, and pointed toward the dance floor can pick up a bit. One on each side, recorded separately, can give you better coverage. You may need an external mixer, pre-amp, or XLR adapter to get the best results.

 

[Allan Black]

Is the 'Basic' the first vid they see? imo a full on dance for a few minutes to set it all up would be better then close ups while you introduce each other.

You'll have to pin the lavs better carefully to avoid the noise of the clothes and if there's too much heavy breathing later.
I don't think you'll need dancing feet sfx, the lighter on their feet the better.

Imo 2 cams should be enough, otherwise editing will get difficult and you'll just be repeating shots, it's the one scene.

The overall look and the curtain is too dark, better lighting plus a few mirrors in back along with assist from the costume dept will all help.

In fact, have you considered instructing while another couple demo the moves? That way you won't get your back showing to the viewers and you can keep your voice at an even level avoiding heavy breathing and distortion, but you'll need careful rehearsals so it all looks pro.

Change the dancers for each lesson to show all ages are doing it? Have you got your music copyright fees covered? When you sell DVDs include a Tango CD, try a tieup with a name band.

Nice website, the Black bck is good but imo you need brighter colored titles?

You might check out Robert Duvalls 'Assasination Tango' for a few editing clues and and some support, in fact why not try contacting him and take it from there, you never know. Hope this all helps.

Cheers.

 

[Alex H.]

Also, make sure that your camera's audio level setting is on Manual and not Auto. Your audio also sounds like it's on auto-level, which will make clothing noises sound much louder.

 

[DavidWL]

Audio from the wireless is overmodulated (distorted)... and it's analog distortion, not digital clipping, which means that the problem lies somewhere between the lavaliere mic and the camera's input. You need to work on gain staging, or aligning each piece of the signal chain to send the cleanest and strongest signal possible to the next piece.

Thank you so much! That helps. I'll check. I knew it didn't sound "right" but I didn't know why. I think I was talking too loud into the mic / the mic was too sensitive. There is a knob on it to control volume, and I'll try turning that down.

The first thing is that you have headphones connected to the camera at all times... especially when recording. If you aren't listening as you record, there's no way for you to know what you've got until it's too late to fix it.

These are free videos we are posting online, and the plan is for us (the instructors/filmers) to do hundreds of them. So we don't have an extra person to listen "real time", but eventually we'll learn how loud we can get. For testing, one of us can listen and the other talk. We did do that, but it's harder to hear then (don't know why).

Start with the transmitter, if there's a level adjustment there (even better if there's a clip indicator, usually a red LED). Since you didn't give the model number of the wireless system, I'm just guessing here. Have the talent speak at "performance level" (in other words, not mumbling or whispering). Bring the input level up just until you see clipping on the peaks, then dial it back just a tad.

Is there any kind of level meter or clip indicator on the receiver? On these cheaper systems, not usually... just curious.

Re: level meter or clip indicator on the receiver
No, there isn't, although there is a "green/yellow/red" series of bars on the camcorder, and we did try to check that. Obviously that didn't help!

We will adjust the sensitivity more however!

Before you buy a second wireless system, you might want to read the "$500 Wireless Question" sticky at the top of the Audio forum. Just sayin'... there are better things out there than the cheap AT VHF systems, and you get what you pay for.

While we could spend a few hundred dollars on audio equipment (maybe up to $400), "good" equipment for 2 people is outside the budget, so more the question is how to best do with lower grade equipment. I think a lot of the problem is the over modulation. We'll also do a little bit of noise cleanup afterwards.

As for the feet of the dancers, a shotgun at floor-level, just off-camera, and pointed toward the dance floor can pick up a bit. One on each side, recorded separately, can give you better coverage. You may need an external mixer, pre-amp, or XLR adapter to get the best results.

Are you suggesting that we use shotgun mics instead of wireless??? I think it would be a problem when we turn away from the camera.

Overall - thanks a ton - especially in the diagnosis of the noise problem, which really bothered me, but I didn't know what to do about it!

Thanks!
David

 

[DavidWL]

Is the 'Basic' the first vid they see? imo a full on dance for a few minutes to set it all up would be better then close ups while you introduce each other.

One thing I am very aware of is the low attention span for online video, so I am reluctant to do this. While we could do more flashy stuff (good for poor attention span), it isn't what we could teach the beginner. Either it is boring or irrelevant. If she were young 'n cute, and we had great camera work, perhaps we could get away with it, but I don't think it is practical even then.

I really want to keep this _immediately_ relevant to the viewer, with as little filler as possible.

You'll have to pin the lavs better carefully to avoid the noise of the clothes and if there's too much heavy breathing later.

Is that why their is rustling? What does it mean to "pin the lavs better" ... tell me what to do (or shoot me a link, etc.) and I'll do it! Thank you!!!

I don't think you'll need dancing feet sfx, the lighter on their feet the better.

Imo 2 cams should be enough, otherwise editing will get difficult and you'll just be repeating shots, it's the one scene.

No, this is for a combination of education (example: Zoom in on feed when she crosses) as well as varying the scene.

Because we are using static cameras and then moving our bodies into the view of the camera, this lets us cut form one camera to another, making time pass more quickly.

Editing is a pain, but I don't think too bad, because once we sync the cameras on the timeline, we can just go from one camera to another.

The overall look and the curtain is too dark, better lighting plus a few mirrors in back along with assist from the costume dept will all help.

We can't really change the curtain (this is the only practical place for us to film). However, we can change the lighting ... what specifically should we do?

We have 3 lights, which can put out between 600 and 2000 Watts (Halogen). Right now we are using 2 (one to the left, and one above and a little to the right). What do you suggest?

Similarly with costuming ... what would you change?

In fact, have you considered instructing while another couple demo the moves? That way you won't get your back showing to the viewers and you can keep your voice at an even level avoiding heavy breathing and distortion, but you'll need careful rehearsals so it all looks pro.

Change the dancers for each lesson to show all ages are doing it? Have you got your music copyright fees covered? When you sell DVDs include a Tango CD, try a tieup with a name band.

Yes, we have. We have done this in a few videos. For example:
http://learntodancetango.com/video/36370976/#section=comment

However, since we aren't paying, and we are doing hundreds of videos, I'm reluctant to impose on people. In addition, we are better than our students, and we want to provide the best model possible.

Is it that bad to talk with my body facing away? It would be the same if I voiced over ... ?

These are free, and so there is no DVD. Regarding music, for our 2nd batch we wrote to many independent musicians and got their permission.

Nice website, the Black bck is good but imo you need brighter colored titles?

Thank you. I like the color, as long as it is clear to read. (Tango isn't really a "bright" themed dance. Also, a brighter read might look garish). But ... do you find the font very legible? If not, then it is definitely a problem.

You might check out Robert Duvalls 'Assasination Tango' for a few editing clues and and some support, in fact why not try contacting him and take it from there, you never know. Hope this all helps.

Cheers.

Thank you - you're right, a good idea to ask (doesn't cost anything!).

Thanks a bunch!
David

 

[DavidWL]

Also, make sure that your camera's audio level setting is on Manual and not Auto. Your audio also sounds like it's on auto-level, which will make clothing noises sound much louder.

Thanks. The audio isn't on autolevel (I don't think), because I set the input mike level on the camera.

When do you hear clothing sounds? Is this like Allen said, perhaps because of the way we put on the lavs?

Thanks!
David

 

[Alex H.]

Re: level meter or clip indicator on the receiver
No, there isn't, although there is a "green/yellow/red" series of bars on the camcorder, and we did try to check that. Obviously that didn't help!

We will adjust the sensitivity more however!

Generally speaking: green is good, yellow is safe but close to the danger zone, and red is a warning that you're about to clip. Theoretically, you want your dialog to average at -12dB. Often, these little consumer cameras don't have dB markings on the audio level meters, so you have to go with some other visual reference mark (hopefully) provided by the manufacturer. My guess is that the line between green and yellow is it.

And again, make sure that auto level control is turned off... the job of auto levels is to keep everything as loud as possible without clipping, but it pretty much destroys dynamic range and makes the background sounds loud as well.

While we could spend a few hundred dollars on audio equipment (maybe up to $400), "good" equipment for 2 people is outside the budget, so more the question is how to best do with lower grade equipment. I think a lot of the problem is the over modulation. We'll also do a little bit of noise cleanup afterwards.

Well, you get what you pay for. As for noise cleanup, focus first on getting your dialog clean and clear and at the proper levels. Do that and you won't have as much noise removal or cleanup work in post.

Are you suggesting that we use shotgun mics instead of wireless??? I think it would be a problem when we turn away from the camera.

Absolutely not. The suggestion was not in lieu of wireless lav systems. It was additional mic coverage - recorded separately - to capture sounds of the dancing (not dialog), but if the dancing style isn't dynamic enough to create useful "feet on the floor" sounds, then there's no reason to do this anyway.

 

[Alex H.]

Thanks. The audio isn't on autolevel (I don't think), because I set the input mike level on the camera.

When do you hear clothing sounds? Is this like Allen said, perhaps because of the way we put on the lavs?

On second listen... that's not fabric noise. It's the hiss of a low-quality signal path with your wireless system (complete with dropouts).

 

[DavidWL]

Alex,

Could you point me to one of the minute:seconds when you here this, so I can listen to? I think I don't have the training to spot it as well as all of you, but I'd like to listen over it a few times, and I might give in and spend more

Thanks!
David

 

[Alex H.]

The hiss is constant... just part of noisy (cheap) electronics.

Dropouts (where the transmission completely fails and all audio disappears momentarily) at :20, 1:42, 2:15-2:20.

And you can hear the hiss get louder as the signal path is weakened (though movement, or RF interference from clogged frequencies) at :44, :50, :55, 1:07, 1:28.

 

[DavidWL]

The hiss is constant... just part of noisy (cheap) electronics.

Dropouts (where the transmission completely fails and all audio disappears momentarily) at :20, 1:42, 2:15-2:20.

And you can hear the hiss get louder as the signal path is weakened (though movement, or RF interference from clogged frequencies) at :44, :50, :55, 1:07, 1:28.

Thank you! I will listen.

Best,
David

 

[DavidWL]

Thank you SO!!!!! much.

While I was reluctant to spend the money, I did, and I am SO!!! glad that I did. The sound is beautiful!!!!!!

Thank you for the time you spent convincing me. I Really, Really, Really appreciate it a ton!

Very Best,
David