App for audio feedback frequency detection?

Question

For curiosity I've started writing a mobile application that would detect when there's audio feedback on the live stage and what's the feedback frequency(ies) so that an sound engineer (SE) can turn down the corresponding band(s) on the EQ. And I'm wondering if anyone else is interested in this problem. I'm asking myself some questions regarding the fight with feedback. Could you please take a look at them and help me?

1. Does the frequency need to be detected exactly or just seletecting its EQ band is sufficient and more interesting?
2. Is for less experienced SEs detecting the corrent band by ear hard (neither fast, not precise) and more more experienced ones a lot easier?
3. Does even for experienced SEs detection by ear get worse if they're tired after many hours of work and/or would they like to confirm his guess by an exact measurement?
4. There exist some general purpose real-time spectrum analyzer hardware units and mobile app. Do you thing a specialized app that would directly tell you the corresponding band on your EQ (in addition to the precise frequency) would be more beneficial?
5. Also there exist some expensive hardware feedback eliminators that automatically adjust the EQ on your behalf. Some people tell that they are too aggressive in killing feedback at the expense of making the sound too dull. Do you thing a more lightweigtht and half-manual thing like a mobile app could serve better?
6. Do you thing that maintaining the list of feedback frequencies that occured during the live performance would be useful? Eg. to quickly eliminate any of those if it occurs again.
7. Is it important that the feedback detection ignore normal instrument and voice tones that do not make a feedback?
8. Is the almost real-time responsiveness (eg. delay of just several milliseconds) really important?
9. Do you think that it is critical that the app just gets along with the built-in microphone or connecting to an auxiliary audio outout from the mixing console is not a problem?
10. Would a phone or a tablet with bigger screen be better suitable for such an app? Or it doesn't matter?

I'd be very grateful if you could look at those question and answer any one of them? Besides you raise your karma I could fine tune the hand-crafted app to your specific needs and provide you with a early beta versions :)

Thanks a lot!

Answer 1

I guess the mixer guy, if he's on the ball is probably not going to use it. Here's my reasoning.

1) If he's got a graphic equalizer to hand then he probably has a bargraph type spectrum analyser. I remember a Klark Teknik one that I used many years ago. It's probably all on a PC nowadays. Same principle though.

2) If he's done a good job setting up the gig then he may well have used a pink-noise source to drive all the speakers and flattened the response already. I think the modern day trend might be to use a fast swept signal generator going thru all the audio bands.

I don't want to pour water on your idea - it could certainly work but it's the convenience factor or maybe the inconvenience factor of using an android phone while you're mixing.

Answer 2

Does the frequency need to be detected exactly or just seletecting its EQ band is sufficient and more interesting?

Given you know the type of graphical equaliser the SE is facing, it would be much more usable to give the band rather than the frequency. But even more usable would be to show both.

Please be aware that with real-time Fourier Analysis it is impractical to detect an exact frequency as the process involves bins which are rather low resolution.

In addition, an issue you'll be facing is that most feedback starts at a low-frequency and quickly becomes high-frequency feedback due to the increase in harmonic content. So feedback detection is often a matter of working out the harmonics rather than a specific frequency.

This is to say that after a short while, the dominant feedback frequencies may be 1K and 2.5K, but the engineer may actually pull down 500Hz based on noticing this low frequency just when the feedback started.

Is for less experienced SEs detecting the corrent band by ear hard (neither fast, not precise) and more more experienced ones a lot easier?

Yes. Inexperienced SE find it hard to recognised specific frequencies. With training, and with some talent, experienced engineers don't even think about it - they hear feedback and know immediately which band to approach on the GEQ.

Note however, that engineers are often aware of the usual suspect frequencies from soundcheck.

Does even for experienced SEs detection by ear get worse if they're tired after many hours of work and/or would they like to confirm his guess by an exact measurement?

Anyone who is tired will under perform. But the effect on experienced engineers is little. Alcohol, on the other, will typically increase their response time more noticeably.

There exist some general purpose real-time spectrum analyzer hardware units and mobile app. Do you thing a specialized app that would directly tell you the corresponding band on your EQ (in addition to the precise frequency) would be more beneficial?

Not everyone can afford or work in a place where hardware solutions exist.

As previously mentioned, feedback detection is not exactly spectrum analysis. So a specialised application would be beneficial.

Also there exist some expensive hardware feedback eliminators that automatically adjust the EQ on your behalf. Some people tell that they are too aggressive in killing feedback at the expense of making the sound too dull. Do you thing a more lightweigtht and half-manual thing like a mobile app could serve better?

Not sure I get this one - will the mobile app have audio in and out and will cut offending frequencies?

Do you thing that maintaining the list of feedback frequencies that occured during the live performance would be useful? Eg. to quickly eliminate any of those if it occurs again.

That depends whether the mic is stationary or not. Most feedback problem are caused when the mic is not stationary. For the former case - it could be beneficial, particularly taking a snapshot of the feedback frequencies from the sound check.

Is it important that the feedback detection ignore normal instrument and voice tones that do not make a feedback?

Normal instruments and voice should not interfere with feedback detection. I'm not sure what you mean by 'ignore'.

Is the almost real-time responsiveness (eg. delay of just several milliseconds) really important?

Several milliseconds is not an issue. Most feedbacks take around a second to build up, and the response of a human being can be between 30 - 500ms (depending on the distance from the fader).

Do you think that it is critical that the app just gets along with the built-in microphone or connecting to an auxiliary audio outout from the mixing console is not a problem?

The built in microphone is likely to distort. Background noise and room reverb could all interfere with the detection process. So a system where the FOH output is connected to the detection device should, in theory at least, work much better.

Would a phone or a tablet with bigger screen be better suitable for such an app? Or it doesn't matter?

Given the context in which it is used, there is a clear benefit for large screens, so a tablet would be better. However, tablets can more easily be stolen, but that's not your issue as a developer.