Fixing voice – too wide DR – using AMM or compression

[volounteer]

One problem we have is speakers that have TOO much DR. They can go from very low to very loud. One was VERY low to VERY loud.
Another is ‘amateurs’ using a wireless dynamic mike and holding it by their waist so they are too low to be heard.
This exacerbates the problem people have with hearing the speakers so they can understand them.

Nothing we can do about those who refuse to use a mike.
Dont know why, but some think they do not need one. Maybe they are afraid of them.

What I want is UPwards compression on voices. Is there a standard compressor on the Qu32 that does that?
What I see is a normal high SPL lowering compressor.

What about AMM??
Can I set some parameters in AMM that would help with the DR problem ?
It would only be working on one mike. Multiple speakers is not an issue with us.

I had thought about compressing the voice early on the channel strip AND using AMM too hoping that would work.
Not sure how AMM really works to see if that makes sense.

I don’t think we have any, let alone enough, spare channel(s) that we could use parallel compression with.
I need to really look at the block diagram and how we are set up to be sure if that could be an option.

I really would like this on at least 3 potential channels: Pulpit, one wearable wireless mike, and one (or more) wireless mike(s) that is portable. Typically only one of those would be in use.

This voice range issue seems like it must be a problem with a lot of AH users.
Seems odd I do not see an upwards compression option in the library.

After fixing the voice intelligibility problem, I also need compression on the outputs to keep the actual audio SPL from being too loud.
This is for church not a rock concert. 85dB SPL typical, *maybe* 90 dB SPL peaks, would be plenty of sound level for the music.

Anyone with suggestions or possible solutions please respond.

[Dick Rees]

Friend…

This forum is for discussion of Qu series mixers, not for getting educated in the basics of sound reinforcement. I suggest you join and post your requests for general audio tasks on the Pro Sound Web. There is only one requirement: you must use your real, full name to participate.

I’ll leave you with one hint:

You need limiting, not compression. Yes, you can set up a compressor as a limiter, hence it being called a compressor/limiter in common audio parlance.

Good luck.

[volounteer]

@Dick Rees
This *IS* about Qu mixers!!!!!!!!!!!!!!!!!

I KNOW the basics of sound reinforcement.
What I do not know is how to make the Qu do what needs to be done.

Does it have upward compression or not?
Would the Qu’s AMM help solve the problem? Insufficient info in documentation to be sure what it can really do.

These are Qu issues for us.

And FYI A limiter compresses the high end.

We need UPwards compression on the low end of sound levels.

The high end is not a problem if the Qu lets us do both the low end as well as compress/limit the high SPL levels.
It is unclear whether their compressor can be set differently for the two different uses.

[Mike C]

Get your terminology straight…when you say “high end” most are going to think
high frequencies and “low end” well bass.

The simple fix is for the “low end” sounds is to have someone turn up the mic channel
when someone speaks soft, granted you can only turn it up so far.

Have fun I think we’re all about done with you here.

[Mike C]

Get your terminology straight…when you say “high end” most are going to think
high frequencies and “low end” well bass.

The simple fix is for the “low end” sounds is to have someone turn up the mic channel
when someone speaks soft, granted you can only turn it up so far.

Have fun I think we’re all about done here.

[Dick Rees]

This is a quest for a technical solution to a human problem. In such situations I have gotten all parties involved to sit down together and come to an understanding of the requirements and responsibilities involved in effctive presntation and participation.
It has been successful more often than not.

[volounteer]

@Dick Rees

Easier said than done. Some guest speakers have their own style no matter what.
Others are not receptive to being told how to do things, or they forget.

Changing someones speaking style is not as easy as you make it out to be.

[volounteer]

@Mike C

Apologies for not noting SPL as the units. Consider this an answer to both your posts.

I thought the context was clear, but apparently not.

The high SPL levels need compressing/limiting to keep them from being too loud.

The low SPL levels need UPwards compression so that the speakers can be understood.

Music directors and audio honchos everywhere claim they get people who say it is too loud and some who say it is too soft.
What they dont grok is that the music *is* too loud on the high (SPL) end, and the voices are too low on the low (SPL) end of the amplitudes (SPL) of the show.

Proper compression both down at the top (SPL) end, and UPwards compression at the low (SPL) end would solve the problems that many churches have. Your simple fix only handles part of the problem.

See pictures of many compression types here:
https://dt7v1i9vyp3mf.cloudfront.net/styles/news_large/s3/imagelibrary/P/Pcomp-02-scales-M.AEL1Ko7.qSUfwPcru.Z3WjThxFeqIa.jpg

http://www.soundonsound.com:
Upward compression increases the loudness of sounds below a certain threshold while leaving louder sounds unaffected. Both downward and upward compression reduce the dynamic range of an audio signal. Parallel compression is just one way to achieve UPwards compression.

[SRD]

Hi volunteer,

Is it not an option to simply set the compression threshold trigger level really low, set a ratio of maybe 10:1 (or whatever works) so all input is compressed. Then use the makeup gain to normalise the volume.

Gate out anything below human voice.

Just my two pence

[volounteer]

@SRD

Thanks considered that. But we have measured up to 60+dB SPL background noise. HVAC mostly. Audience could increase it some.
One problem we have is the wide DR of many speakers (people orating). Amateurs are bad but some of the ‘pros’ can be even worse. And gating could cut out low speakers intermittently

What we really need is upwards compression so we can lift just the voices and let us mildly compress the high (SPL) end so it still sounds good.

I think I need to invent a mike that has built in AGC to help improve communications and intelligibility for users like us.
Actually I did. I guess I should patent it real fast:)

[Ryan]

Turn on the HVAC before the sermon, turn it off during the sermon, then turn it back on again after.

If your microphone is picking up 60+dB of background noise, then upward compression will also make whatever background noise is getting picked up louder. Unless you’re saying it will somehow make just the voices louder and ignore everything else the mic is picking up?

[Ryan]

In the soundonsound article that you linked to, there are 3 points that I think are relevant:
1. It mentions that finding a physical device that can do upward compression is rare
2. It mentions that parallel compression is used to achieve the same goal as upward compression
3. This direct quote about upwards compression: “However, if you spend a few moments thinking about the idea of genuine upwards compression, you’ll realise that it wouldn’t actually be a lot of use in real-world applications, because it would probably raise the noise floor to wholly inappropriate levels. For example, take an upwards compressor with a gentle 2:1 ratio and a threshold at -20dBFS. If the ambient noise floor of the recording was at, say, -80dBFS, the act of compression would drag it up to an appalling -50dBFS, which probably isn’t really what you’d want!”

[volounteer]

@Ryan

The A/C is controlled by a thermostat. The temperature is set to maintain a steady-ish comfortable level.
Turning it off when the room is full of people is a non starter.

I hoped that we could gate the signal if needed and if the speakers (people talking) were loud enough that we could minimize the problem.
60db is what I measured at the point I listened. I presume it would be similar on stage. But the mike being directional may not pick up as much as the SPL meter. Maybe it is quieter on stage. I know it is touchy as when they boost the gain we run into feedback issues so there is limited fader changes that are feasible.

The meter was not lab calibrated so is possibly high. We do not seem to be amplifying noise now but are NOT amplifying the speakers(people talking) enough when they are low. What we need is the output from the speakers (devices) on the ceiling to be about 15 dB above the background noise for maximum intelligibility.

From what I can read upwards compression would help us improve the intelligibility of the people talking into the mikes and avoid the complaints about things not being loud enough.

[volounteer]

@Ryan

wrt 3 points

Rare but not non existent. I could do it in software if AH had an SDK.

Parallel compression is the usual way to do it hence no big need for a device.
But an ap is easy and cheap.

Noise is always an issue as is Gain before feedback.
Combined with the S/N needed for intelligibility the system must be set up carefully.

I had initially computed we only needed 2:1 which is gentle.
Gate above the noise level, add 10dB make up gain and we achieved the optimum S/N for our level of BG noise.

As to being needed, I disagree. Most churches need it as they all seem to have complaints that the sound is too low.
What they mean is they cant understand the preacherman. The audio folk say but half the people complain it is too loud.
They are ignorant andor confused. The music IS too loud. The voice is too soft. They need to apply upward compression on the voices and normal compression on the loud end. Then everybody in the audience would be happy.

[garyh]

Get yourself a mic preamp with thresholdable agc. Some are even powered by phantom. Set and forget.

[Author]

Posts