Help Me Get Good Sound Recording With My Video

Hello everyone

Im using a Axis Q1604 camera mounted in the back of a church about 50 feet away from the Pastor. Using it for live Streaming. Im not getting very good sound with the internal mic with dB all the way up. So im looking at Louroe and placing a varity of verifact mics around the church. Im wondering if there might be better solution that is easier. Their existing sound setup is not really an option due to the fact they have 3 separate mic systems not always being used. They don't have a central Mixer.

Wondering if you guys have done a setup for live streaming and how you handled the sound. Im donating my time and mtrls so price is a factor.

Current Setup

Camera Q1604

Axis Streaming Assistant

Adobe FLME 3.2

LiveStream

Thank you for any help you could provide.


Hello Lee:

I'm sure you're aware that mic build plays a big role in the range of fidelity it picks up. In that way, most electret condenser types, and most 'surveillance duty' external mics are built to pick up the midranges (ie: human speaking voices), not low frequency bass notes or music.

What I am suggesting: even if you go through the expense of adding an external mic to your camera, it probably won't sound as clear or have the range of the ones coming through a soundboard, especially if the room is big. The closer you can get the mic to the source, the better the pickup will be.

With that said: connecting a mic to the Q1604 is straightforward. The backend of that camera looks like this:

That pink 'IN' port supports any 3.5mm mini-RCA microphone. I have used this mic in a meeting room application directly connected to an Axis camera in this way:

I am not sure how the streaming services you are using will impact audio latency or sound quality, but I know that audio inputs from the camera will be noticeably improved over the stock internal mic.

Does that help?

Hi, Lee and Brian!

I'm responding to Brian here because although I agree with alot of what Brian is saying I would have to disagree with the mic choice for a couple of reasons, although I do own the mic myself and think its 'ok' in certain cases.

1. The DS70P is a stereo mic and your input is mono and the way axis treats stereo mic ins is to discard the right channel. So you are paying for 2 mics and using one.

2. The pickup pattern is uni-directional stereo, not ideal esp. if only one side works. If you are looking to pick up a speaker from a distance, you want a more directional mic (think football sideline mic, not quite but same idea).

The primary issues with speech clarity are going to be mainly dealing with reverb wash. A sound survey would identify the best placement of the mic and suggest the optimal setup, tho a couple basic rules wil always be true.

Use a highly directional mic, like a shotgun mic if it is in the budget.

Avoid the corners(unless using a boundary mic); standing waves create various frequency build-up called modes which are worst in the corners.

Use a low-cut filter or a mic with reduced lf-response.

You say there is no total mix you can tap, but I assume there are speakers which amplify at least the voice component?

One effective, although admittedly low-tech, idea is to simply setup a directional mic three or four feet pointed directly at the cone of the speaker. This will enhance your audio stream quality greatly!

One mic with which I have direct experience with Axis is the Sony ecm-cz10, its super-directional and works great with speech. It also comes with a shot-gun like tube that you'll want to use.

P.S. I'm assuming that its not possible to use a wireless lapel mic here(best), also you could get the stereo mic and get a y cable to combine the inputs, tho im not sure how it might affect the phantom power...

There are other things that affect audio quality(codecs, vbr etc), obviously, I'm just speaking to mics and placement.

Let me know if you have any questions.

C

I'll stand by my recommendation because time and cost is a factor.

I am not an audiophile, and I have no doubt what you are writing is correct. However, the mic I recommended cost $60. Once we start discussing sound surveys, preamps, shotgun and boundary mics the cost quickly rises!

Pointing a directional mic at a speaker may be a good, pragmatic solution. However, feedback could be an issue, right?

Correction the mic(s) your recommending are $60, the one I'm recommending is only $80 and is IMO vastly superior in everyway except its stereo-ness which cannot be used. I own both. I didn't say anything about preamps.

Just this mic direct.

Feedback is not an issue because the the streaming mic does not 'feed' the speaker.

Roger that. Thanks for the recommendation!

If the camera is in the back of the church, I really would get a decent directional mic. A shotgun mic (I used the AT897 back in my audio days) should do the job. You'll have to get preamp with phantom power, though, like the RDL EZ-MPA1, to power it, and also because I'd not trust Axis' internal mic preamp for this sort of streaming.

If you get something omnidirectional you're going to be picking up coughs and whispers and you'll lose a lot of clarity.

I agree with Ethan, use a professional audio mic. I undersand cost is a factor but don't expect much from using the audio through the camera. You can get a small mixer/pre amp for $50. I have used one from Behringer that I bought from a local music store. Back in the day, I got into a niche of police interrogation rooms. I used a Biamp, mix-minus system with echo cancellation. I was the highest price but once I sold one, the referrals rolled in.

One other consideration that hasn't been mentioned is that, if you put a mic at the back of the church, there WILL be a time-delay factor and audio will seem to lag slightly behind the video - if your camera/mic position is, say, 100' from the stage, there will be a 300ms delay in the sound arrival - enough to be noticeable and probably a little disconcerting to those watching, but not enough for most people to quite put their finger on WHY it's disconcerting. The closer the mic can be to the stage, the less this will be a factor.

Problem with a shotgun mic is, like so much in surveillance video, you're trading off one problem for another: the more directional the mic, the less outside noise, but the more sensitive the aim becomes. You could point one right at the pulpit, but if it's TOO selective, you'll start to lose audio as the speaker walks around. It's like having a tight zoom on one spot vs. a wide shot; detail vs. coverage.

Anthony is on the right track here, your best bet may be to implement a small mixer of some kind, and tap off their existing separate mics, since people are presumably already familiar with the operation of those systems anyway (one less thing you have to train on).

The other option, though not the cheapest, is the aforementioned PZM (pressure zone, or boundary) microphone directly on or above the stage. They're very good at wide-range pickup without picking up "direct" noise (like people stomping, if it is actually ON the stage), they don't suffer from "proximity effect" (where up-close sources get "boomy"), and pickup pattern can be tuned using any sound-absorbing material like a small rug. Back in the day, Radio Shack used to sell Crown factory seconds (under their own Realistic brand) for around $100; we used a couple of them in the my church and they were fantastic for all kinds of things.

Thanks for all the replies. I thank you for all your input and suggestions. Found that they had mixer under table that we moved all there mics to that unit. We got it working on there Peavey XR684 im going out on the Monitor output. It's the only one i could get to work. It much better then before but i had to turn the input gain all the way up to 12.5 db on axis audio settings. And also had to turn up there moniter volume setting all the way up on the mixer. Not sure why? Come to find out they have some audio issues with that mixer already. We are fairly sure that that unit is going bad. So now we looking for a new mixer for them. Might just have to have Audio expert come in and help us out.

Im am getting a delay in the audio as well not sure if that from streaming or what? But they don't seem to match.

@Matt:

100' from the stage, there will be a 300ms

1. I know Canada is cold but where I live sound travels at 1 ms/1 foot, so 100 ms. (Ok 1.12 ms, but every studio hand knows it as 1 for 1). If point #6 is followed there might 10 ms.

2. As for boundary mics, Crown's are the name though placement is not easy and takes trial and error or sound survey, and last I checked they require seperate power sources.

3. Sure tapping in to an existing mic/ submix would be great but the OP seemed to think this was problematic. But of course if he could do it [note: now it looks like it may be possible! Yipee!]he probably would have never posted.

4. As for the shotgun mic be too directional, you are thinking like a video guy, audio has a much, much wider rate of dispersion so much so that at 50 feet it would envelope the whole stage. Perhaps Prof. Carl has a textbook he has written to back me up.

5. Proximity effect is unlikely to manifest in any of the proposed solutions due to a shortage of .... proximity.

6. Anyway, by far the best option if price/time is critical is to get a mic, any decent mic close enough to a speaker to get reverb free voice.

Which leads me to the twice maligned ecm-cz10, with wave tube included. For under a hundred bucks I feel its a clear winner.

I know Canada is cold but where I live sound travels at 1 ms/1 foot, so 100 ms. (Ok 1.12 ms, but every studio hand knows it as 1 for 1). If point #6 is followed there might 10 ms.

Oops, right you are... been SO long since I've worked with audio theory, I forgot the 1-for-1 rule and trusted a handy online calculator that seems to always caluculate for meters no matter what you tell it.

Anyway... it's still enough of a delay that it would feel "off" to most people.

No more "off" then if you were standing where mic is, so actually it depends on the FOV of the camera right?
Another rule that I have heard sound engineers say is that never let the audio run ahead of the video.

Our mind has a much harder time dealing with it (as it should). Its subjective but I've heard audio should be no more than 75 ms behind and no more than 25 ms ahead.

C

Not sure what the FOV has to do with it... point is, if the mic is too far away, the audio will be slightly behind the video, and it will seem weird to people watching. Heck, watching a video in VLC, if I offset the audio by more than about 20ms either way, it seems weird - people are talking, but their words don't QUITE match their mouths, almost making it seem like an overdubbed Bruce Lee movie.... or something like that.

Not sure what the FOV has to do with it...

So Matt when you are standing 50 feet away at a big church do you often complain, "Wow, the audio seems off" ?

Of course not because your mind expects the audio to lag based on how far away what you are seeing is, i.e. your FOV. Now on a closeup the lag might seem strange but thats because your FOV has changed. It's all part of psycho-acoustics. In the same way the camera makes us think we are a certain distance from the sound.

I bet you are the kinda guy (like me) who wants his mono sub woofer in the center of the room even if the manual says it can be off to the side. It really doesn't matter. Weird psycho-acoustics.

Try again in vlc (blind test if possible) setting the audio to lag 20ms, (on a known in-sync track of course), watch for 10 minutes without a/b'ing anything, you might think its ok then. If not it sucks for you like perfect pitch can! Out of sync stuff bothers me too but not til >50 ms

Cheers!

Why not add a separate central mixer dedicated to the streaming? Music industry based mixers from many manufacturers are inexpensive and work very well.

Also make sure you are using AAC (Advanced Audio Coding) compression which is capable of higher fidelity. On most surveillance-grade cameras, the only choice is G.711 compression (technically compansion), which is only capable of VOIP (telephony) fidelity. On this camera you do have a choice, but I don't remember the default.

Also use the highest sample frequency (16 kHz) and see if that makes a noticable difference in quality.

Thanks

It is AAC. But I changed the frequency to 64. I will put it back to 16. It didnt seem to have much affect.

On their data sheet it listed 16 kHz as the highest. Higher might sound better, so 64 might be better. The bit rate will be higher, but nothing compared to the video stream.

Don't expect a broadcast-quality image. There are big differences between cameras made for surveillance and those made for television.