Feedback reduction mic

This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
I am speech impaired which the speech volume is extremely low as a side effect of thyroid operation. I am part time teacher now. I tried several portable pa, even some expensive, none of them is satisfactory. The feedback are horrible for I need to raise the gain of mic. to very high level.

I am exploring the possibility to diy a feedback reduction mic. pre-amp. In some cell phone and bluetooth head phone use dual mic. solution to cancel surrounding noise. Not sure this is applicable to my need and if the components are available.

Any suggestion are welcome.

Thank you.
Before fancy solutions (which *may* work but involve more experimenting) , and given the urgency, let's go the tried and true methods.

1) buy a *good* headset microphone, not the cheap computer type but a PA quality one, and set it right in front of your mouth, to maximize sound pressure at the capsule.

2) put the speaker(s) as far as possible; in preference at the side walls , pointing at the center of the classroom, and 1/3 the room length, putting most of their sound in the pupils area and away from you.

3) there are some feedback cancellers (sweepable notch filters) which can help you tune any remaining feedback out, but complying with points 1 and 2 should be more than enough.

4) maybe getting an Acoustic Guitar amplifier and setting it on a table some 5 or 6 meters away, pointing at the center of the room, is an option.
They are light, portable, have a Mic XLR input, sound very good, and already have a sweepable anti feedback notch, although only for the low/low mid frequencies.

feedback ocurs when the sound from the speaker gets back into the microphone. So the first step is to set up the speakers to minimize that. When you face the audience, the speakers should be further forward than you are - closer to the listeners. And they should be at an agle so they do not face towards you. If you can look towards the speaker and see its face, then its sound can get to your mic.

Noise cancelling and feedback cancelling are maybe related, but not the same thing.

However there are commercial feedback killer units that you can add. ANy system that feeds back tends to do it first at certain frequencies the system is more sensitive to. Feedback killers detect this and reduce those particular frequencies. Some of them work pretty well. There are also some feedback killers that change the frequency very slightly - a pitch shifter. So the sound going into the mic comes out the speaker at just a little different pitch. so the feedback loop can never form because each trip through the system results in something different. That is not my favorite, but for spoken word, such subtle shifts are less noticable than in music.

Look for example at the Peavey Feedback Ferret.

But work on a good mic and better speaker arrangement first.

You can get a pressure gradient anti feedback microphone.

Sound quality is not good but they work. The principle is
two capsules wired to cancel the pick up of external noise.

By speaking with one of them nearer the mouth you get
a differential signal that you can amplify without feedback.

The things they do for mobile phones and bluetooth headsets
are way beyond any sort of DIY preamp, its very complicated.

Military surplus should be a good source, they are the sort of things
used e.g. in helicopter radios to the cut background noise for speech.

rgds, sreten.
Last edited:
Noise cancelling mics are generally held a fixed distance from the mouth, either by a fairly rigid head mount or the mechanical structure of the mic itself (see the STC/Coles 4104)

Feedback rejection can be done with equalisation (although the frequencies notched out always seem to be the ones most important for comprehension), by distance factor (don't ever try using lapel mics; not only do they need more equalisation than is generally available in a simple rig, but they're a long way from your mouth; a handheld or proper headmount work far better; I've even an ex-military throat mic for extreme conditions where quality isn't critical (carbon granule mic, so quality really doesn't need to be important).

Electronically noise cancelling systems for use in high noise environments - frequently helicopter cockpits – exist, for communication frequencies, generally come attached to noise cancelling headphones (leaving you isolated from class responses) and are fairly specific about powering/interconnection.

One trick I've pinched from the BBC in the sixties was to put a slight pitch shift in your signal chain (back then it involved multi ring modulators, crystal oscillators and single sideband filtering, technology developed for one of Stockhausen's 'mikrophonie' pieces; nowadays any cheap effects unit will give you the same result at the touch of a button). That way, the feedback can't stabilise into a long, continuous 'weeeee' but gives a sort of 'bouoing' effect as it is pushed out of audible frequencies. Your voice sounds a bit funny, true, and you couldn't sing through it, but comprehension is reasonable.
Thank you for valuable suggestions.

Better setting of loudspeaker is applicable, if I teach high school. I am teaching primary students and walk around the classroom frequently.

The pay for a part-time teacher is very low that can't afford expensive equipment. I send e-mail to resigned. But I still interest in making a feedback reduction mic.

Could anyone suggest a circuit?
Is possible to make an frequency shift android app.?

******** EDIT *******
Find out this DAF Assistant app.
DAF Assistant implements Delayed Auditory Feedback (DAF) and Frequency-shifting Auditory Feedback (FAF) techniques that are known to help people with stuttering to speak more fluently.
DAF Assistant can be used by people who stutter to control their speech fluency, slow down speech rate, increase confidence level, and develop good speaking habits.
Last edited:
Weiyan .
See if you could incorporate a product like the Beheringer Fedback Destroyer in your system.
This will dymamically add notch filters at problem frequencies and operates automatically as you create different feedback peaks. You can also assign some of the filters to EQ the sound to suit your limited bandwith requirements.
They are cost effective and cheap secondhand.
I use one for PA monitor use - 2 EQ frequencies are fixed to suit the monitor speakers the rest automatically kill any resonant peaks - Very effective and they need no user intervention.
Kind Regards
Martin (Xoc1)
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.