Sound Quality on Live Stream from ATEM Mini Extreme

We have an ATEM Mini Extreme that we use to stream two live services on Sunday morning to YouTube at 1080 and love it!!

The input to the live stream comes from an older model Yamaha LS 9 mix via an XLR out of the Yamaha and convert that to a 3.5 input into Mic 1. We are hearing from those watching the stream that it is dropping off at both ends, highs and lows. We also record the output to a portable hard drive and those listening to the playback of that say it sound the same - dropping highs and lows. In addition, we also record the audio of the service via a USB port on the Yamaha board and that sounds much better.

In looking at ATEM control, it looks like Mic 1 and Master (only audio input used) are set the same and are the original settings. Our audio guy says there must be some high pass filter or something in the ATEM that is causing this problem and we should just buy a newer version of the ATEM. In looking online for info on using an ATEM with a LS 9 for live streaming their does not appear to be much. Looking at live streaming using a Behringer x32, there is a lot of information. Obviously there are a lot of parts and pieces involved in all this.

Does anybody have any thoughts on this?? Always open to new ATEM but would hate to get one and find out the problem is still there.

Thanks in advance. Ed

We do the same thing for our services. We found the balanced out from our board was really hot going into the ATEM. We had to turn the gain on MIC1 way down (about 9'oclock).

We also found it best to set the ATEM to 1080p30 for the best live stream. This gives an average of about 5.5-6Mbits/sec.

[Gary Adams]

Hello Ed. I would like to offer some things to check. Like the above post suggests a typical balanced output from a console may be +4 dBm while the ATEM is -10 so there will need to be a 14 dB adjustment necessary. Most importantly though a typical console will be balanced while the ATEM is unbalanced. An incorrectly wired cable can cause issues like you suggest. While there are several solutions to this I recommend a transformer based isolation device in line. There should be no frequency issues when it is connected properly. Definitely no need for a newer switcher. One last thing to be sure. Make sure nobody has enabled the audio equalizer on that channel at that is capable of doing exactly as you describe. I hope this helps.

Regards. Gary

Why not use any embeder ? AES/EBU or SPDIF is much better , then analog ..

Why not use any embeder ? AES/EBU or SPDIF is much better , then analog ..

• Additional cost of an audio embedder
• another device that can fail
• no audio delay option on HDMI inputs, audio console must provide output delay on mix bus
• when configured correctly, the analog inputs are good enough for most applications

I dont like 3.5mm jack .. XLR or BNC .. That is my way ..

The input to the live stream comes from an older model Yamaha LS 9 mix via an XLR out of the Yamaha and convert that to a 3.5 input into Mic 1.

Feed your camera (instead of ATEM's Mic in) from the mixer via XLR.

I would think the only issue with going from a balanced out to single-ended input over a short distance is added noise. Noise could come from a ground loop or induced EMI. I doubt the impedance mismatch over a few feet is a factor.

Also an out from the board is higher than mic level so there's a greater signal-to-noise ratio. We use a 6' dual XLR to 1/8" stereo patch cable that works fine as long as the gain on MIC1 is turned down.

Otherwise both active and passive balanced to unbalanced converters are available. I like the suggestion from a previous post that you could use a camera or HDMI audio injector to convert audio to digital and use a camera input to receive it into the ATEM. At this point I guess you would be comparing quality of A to D converters!

Thanks for all the thoughts!

Couple of clarifications. We use 3 P/T/Z Sony cameras,2 go pros (one in a drum cage and one with a side shot of the stage) and an older Panazonic camcorder that provides a shot of the keyboard at a keys level. All audio input comes from mics on stage, in the drum cage, instruments and a couple of "house mics."

The reason we use a mix off the board to provide the audio to the ATEM is to get a separate mix from the room mix that comes in on Mic 1. This also allows the audio engineer to adjust that mix on the Yamaha which he monitors in addition to running the house mix. Yes, ideally, we would split all the feeds into the Yamaha before they get there and send the "splits" to a separate sound board. This would require an additional audio operator, additional audio mixer, costs, space, etc.

I ran what was suggested concerning the difference in Db levels and the balance/unbalanced comments buy or sound engineer and he said he was aware of these and they have been factored in.

His thinking is still that "somewhere" in all this is some sort of "filter" or setting, either in the Yamaha or the ATEM that he has not been able to find that is causing the issues. I can run sound but by no means an audio engineer.

From what I have been able to find, this practice of one board/one operator works with most all the How To's related to a newer board like the Behringer x32. One on x32, there was debate whether it was best to use an aux out pair or a matrix mix.

Keep on with the ideas and suggestions! Thanks so much.. Ed

Hi,
We use nearly the same setup (LS9 + Atem mini) without any problems.
For a test: LS9: prepare an aux mix (prefader, pre eq) for the Atem Mini. Input could be a cd-player or something else. Control this aux mix with headphones. Result sounds ok? If not, the problem is in your mixing desk.
Next steps for the Atem:
1. Control that the 3.5 mm input is set to line level (in the Atem Control software)
2. Check that there are no EQs are enabled (mic in and master)

In our case
a simple XLR/female to 3.5mm works 80% (hum)
a cheap headphone amp with XLR input (Behringer P1) works 95% (hum)
a Radial SB-6 works 100%

Greetings, Thomas

I suspect the meters you see in the audio pane of Software Control are post-fader, not post-pre. So if you are clipping the preamp you can't easily tell.

The way I saw this was the meter levels didn't vary with the incoming audio. What was happening was the preamp was pegged to maximum output and the meter showed just the setting of the fader.

My approach was to leave the fader at unity and turn down the preamp gain knob until I saw the levels start to vary. For my board's balanced out this meant the gain knob was at 9 o'clock.

A bit lost on why people keep pointing to gain levels, as gain being too high would resulting clipping rather than filtering (loss of particular frequencies)?

To the OP: you may have already accounted for this, but the first thing I would suggest is to verify that the mic input you are using is configured as dual mono and not stereo - by default you get one fader that controls the left and right channels of that input together; you need to split those, center-pan the left side, and mute the right side, due to the balanced signal being sent out of the mixer.

If that is already how you are set up, or if it does not help, you may be able to further narrow things down by finding some other device with a similar 3.5mm input and plugging the audio cable into those to determine if there is a matching filtering effect.

You could also try putting something like an ART CleanBox Pro between the mixer and the ATEM (mixer -> balanced input, unbalanced output -> ATEM), though that is just a guess based on limited data and may not help either depending on where the problem is.

Hi Ed,
we use the same configuration for live streaming and thus solved the audio problems due to the balanced output from the audio mixer to the stereo input of the Atem mini Extreme.

As already suggested by others there are various solutions to solve the same problem, but this is one of the simplest and cheapest.
We simply made a 2mt long cable with the following plugs wiring:
XLR - 1 = ground, 2 = both white and blue terminals, 3 = not connected.
JACK 3.5mm - Tip = blue terminal, Ring = white terminal, Sleeve = ground.

XLR-Jack.jpg (270.18 KiB) Viewed 2398 times

Since we use only the +180° phase pole of the balanced output signal from the audio mixer this becomes unbalanced dual mono and can introduce some noise, but using a short cable with a good shielding we have never had any problems.

Make sure you set the MIC analog audio input of the Atem mini Extreme you use as “Line” in the audio settings panel.

Screenshot 2023-03-01 alle 11.35.47.png (378.42 KiB) Viewed 2398 times

I hope it can be useful.
david

Thanks to the previous post I discovered the MIC inputs can be set to 'Line'. This means the gain doesn't have to be turned down so much when driving it from a board balanced out converted to single-ended. I tried it today with our setup and the gain was fine at 2 or 3 o'clock.

As far as the OP question- dropping both highs and lows is hard to figure. The EQ in the ATEM is flat by default. Usually anything on a sound board can be solo'ed into headphones. That would narrow it down to inside the board or inside the ATEM.

The balanced outputs of the audio mixer sends the same audio signal on two wires over the XLR cable, but inverts the phase of one signal by 180 degrees, (to keep the audio clean from external interference).
If the analog audio inputs of the Atem mini were also balanced there would be no problems and in case of external noise introduced in the signal this would be cancelled, even using very long cables.
Instead, as Gary has already mentioned above, the external audio inputs of the Atem mini accept two separate unbalanced audio signals and for this reason if they are connected directly to the balanced XLR output the result would be "0" with the cancellation (almost completely, but at least a strong dynamic reduction) of the signal itself.
So it is sufficient to use only one of the two signals arriving from the audio mixer (it is indifferent to use the one at +180° or -180°) and connect it to both channels of a single Atem input to obtain a correct dual mono connection on each end.
In the same way it is possible to respect stereophony by connecting only one signal (with the same phase) of the two balanced outputs of the master right and left channels of the audio mixer, one to the Tip and the other to the Ring with the ground/shield in common to the Sleeve of the 3.5 mm jack.

There is no need to mess with a fancy cable.

You can split the input from within the ATEM software and configure it as two mono inputs so that the left and right channels are independent.

At that point, the inverted signal is on the "right" channel and can simply be muted to prevent cancellation.

Pan the "left" channel to center and you should have exactly the same thing that you would get with a custom cable.

It works, but still send a problem to the Atem mini and then solve it via software.
The fancy cable makes a correct connection without having to solve any issue.
It takes a minute to weld and is cheap, just two plugs and 2m cable to get rid of thoughts.

Wanted to circle back and provide an update on what worked for us. We installed a Tech Digital HDMI Audio Embedder and it has fixed the issue!! Everyone is very happy!!

Thanks to all those that responded!! great group and wonderful resource Ed