Connect with us

Enhanced hearing

Discussion in 'Electronic Design' started by PPP, Oct 30, 2006.

Scroll to continue with content
  1. PPP

    PPP Guest

    Good morning!

    I'm trying to make an enhanced hearing circuit (kinda like Superman).
    For my conceptual design, I read some of the tips I've gathered from
    various postings in this group and also from various readings. So far,
    I've come up with a crude concept:

    - Audio input coming from a small microphone (electret-type?)
    - Output from microphone is AC coupled to a voltage follower / buffer
    audio op amp
    - Audio op amp buffer is single supply so i will need a virtual ground
    in one of the input pins that will be approximately 1/2 of the supply
    - From the two steps above, I'm assuming that I won't have any control
    of the DC level bias from the microphone output

    - My buffered signal would then be split two ways... i'm assuming i can
    just connect the buffered output to two different op amp inputs... part
    of the signal will be sent to a level detector and the other part would
    be the actual hearing enhancement gain op amp

    - The level detector part would probably consist of an active rectifier
    with a unity gain with its output being sent to a comparator.... the
    output of this comparator would somehow interface to the gain op amp so
    I can adjust the gain if the audio level input is too high... i saw an
    active rectifier circuit in the Arts of Electronics book, so I think I
    can use that one.

    - The gain part would consist of an audio op amp with a variable
    resistor for gain adjustment... but then this is where I get stuck. For
    the gain section, I do not know how I can amplify my audio signal
    without amplifying the DC component.

    Since I do not have that much experience in electronic design compared
    to others here, I was hoping maybe you, the experienced ones, can help
    me answer the following questions:

    1). How can I amplify an audio signal correctly without clipping the
    output if I use a single-supply configuration?

    2). How can I interface the level detection portion with the gain
    portion? (I was thinking maybe I can use some sort of "voltage
    controlled resistor" if there is one as part of my gain feedback

    3). What parameters would dominate my choice of parts for this type of
    application? (I'd like this to be battery powered, so I'm assuming I
    need some sort of low voltage, low power op amps. But I do not know how
    to spec out the microphone and headphones for battery applications.)

    4). How many questions can I ask before you guys/gals get sick of me?

  2. Not going to work.

    Your ears can already hear way down low. Problem is in almost any
    place except the middle of a desert, with no wind, there's a constant
    background noise level. If you amplify the sound you're just going to
    bring up the background noise. And you're actually worse off, as your
    typical microphone and earphone doesnt have the dynamic range of your
    ear. It's hard to compete with a billion years of evolution.

    If you want this to work, you'll need a very directional microphone,
    like take an old 3-foot fiberglass snow-disk and place the microphone
    at the focus. Works really well.
  3. Rich Grise

    Rich Grise Guest

    I was thinking, maybe an 8" dish cut in half, with one half on each side
    of his head, their focal points at his ears.

    Then I thought, "What a Mickey-Mouse arrangement that would be!" ;-)

  4. PPP

    PPP Guest


    That was actually part of my question... the specs for the microphone.
    I've read about others accomplishing this feat using directional
    microphones, but I don't know much about them. I've also read of other
    setup where they use a differential input to cancel out the noise, but
    I also don't know much about them either.

  5. They're called "hearing aids".

    But as someone points out, when you start amplifying to hear the distant
    and weak signals, you also amplify the closer and louder signals. So
    if you want to hear that older guy barely speaking next to you, when
    you crank up the amplification the frogs in the pond outdoors are far

    Of course, hearing aids aren't quite so much about making things loud,
    as they are about selective amplification, to compensate for hearing

    The details of your "bionic ear" really are too micro, they deal
    with fine detail that ignores the bigger problems.

    You've overlooked the selective nature of "stereo" hearing; having
    two ears really does help to "tune out" the unwanted sounds. The
    head does help.

    And you need to deal with coupling the amplified sound into the ear.

    Some sort of selective and variable filtering might be useful,
    so you can get rid of those obnoxious frogs (though at the expense
    of a slot in the frequency range for the other sounds). The "Bionic
    Man" would have it done all with computer, so there is some algorithm
    to get the sounds you want and get rid of the unwanted sounds without
    much fussing by the person using it.

  6. PPP

    PPP Guest

    I also read about hearing aids using DSP-based devices that offer
    selective frequency amplification, but what I really want for now is
    just some sort of "sound" telescope that can amplify any sound.

    Anyway, the application of this device is somewhat silly. I was
    thinking of creating a tone generator and attaching it to another
    object. I would throw this object or hide it in the bushes and use the
    "enhanced hearing" device to detect the location of the tone. It's for
    some game that I thought about. Of course, if I know the tone's
    frequency, then I can just use some form of tight bandpass filter to
    get rid of other frequencies.



  8. Bob Eld

    Bob Eld Guest

    If you want to do something more advanced than simple analog circuits,
    parabolic reflectors and the like to enhance sound, consider using DSP,
    Digital Signal Processing to accomplish your goals. One thing you could do
    with DSP is frequency shift so that you can hear bats and ultrasonic insects
    and other interesting things either above or below the frequency range of
    human hearing. Another thing is to selectively amplify only certian
    frequencies or certain sound patterns to increase the aural acuity. Still
    another is noise shaping and filtering to increase the signal to noise
    ratio. Furthermore it is possible to do correleation techniques on very weak
    repetitive signals to pull them out of noise. There's many other neat things
    that can be done with DSP. If this is for a class or to learn somethng more
    or less at the state of the art, I suggest you look into it.
  9. Rich Grise

    Rich Grise Guest

    He should just make a functioning pair of Mickey Mouse ears. ;-)

  10. Think about signal to noise ratio. You want to reject unwanted sounds
    based both on their frequency and direction to source. If you can
    characterize the kind of sound you are interested in, there may be
    additional techniques you can use to select them.
    Three. That last one put you over the limit. ;-)
  11. It should also be possible to build a "Yagi" microphone:

    A string of microphones spaced an integer wavelength distance. Amplify the
    signal from each microphone and then add up the waveform with delay lines so the
    same point on the signal is added to the signal and different signals average

    The effect is to amplify the signal from one direction.
  12. Phil Allison

    Phil Allison Guest

    "Frithiof Andreas Jensen"

    ** Bollocks.

    ** A group of identical cardioid mics all mounted in a line, with outputs
    combined, has many dB of gain and strong directionality over a wide
    frequency range.

    ........ Phil
  13. Joel Kolstad

    Joel Kolstad Guest

    Wrong term... you're thinking of a phased array, where (at least to a first
    approximation) radiation/pick-up is modeled as the superposition of each
    antenna's standalone output/input.

    Yagis place the elements close enough together to get significant coupling
    from one radiator to the next.
  14. Won't work very well. The Yagi is a narrow-band device. That's fine
    at radio frequencies as often you're just interested in a small band of
    frequencies, less than 10% wide.

    it's a narrow band device because it requires the reception elements to
    have different phase shifts at the center frequency-- that's how it
    acts directionally.

    But audio freqs span many octaves, much wider than a Yagi can handle.

    Plus the dang microphones would have to be on the order of half the
    audio wavelengths in size-- not too practical below a few KHz.
  15. Rich Grise

    Rich Grise Guest

    No, he's not talking about a phased array, he's talking about a Yagi.
    Or maybe more accurately, a Yagi-analog, or quasi-Yagi. The spacing of the
    elements is less than .25 wave because of those coupling effects that
    you've mentioned (in the electrical one), but the effect is the same: A
    particular wavefront comes along, and hits each mic in turn. Now, the mics
    don't reradiate, so you'd either put them at 1 wavelength or 1/2, where
    every second one has its output inverted.

    That would be very directional, at one frequency, but the signal would
    be swamped by ambient noise.

    Yesterday or so, I tried to search for the "shotgun microphone" that I
    saw a construction project on in the '60's - it's composed of a bundle
    of aluminum tubes, one 36" long in the middle, then in a hexagonal pattern,
    35", 34", 33", and so on, until you finally have two 1" pieces to complete
    the hexagon. Then, they took an ordinary funnel and big flat mic element,
    and it was done.

    I hear that one's not only highly directional, but has enough bandwidth
    to understand people talking.

    Good Luck!
  16. Rather than aluminium I believe fibreglass would be lighter and have some
    damping. Apparently they are a surprisingly effective directional mic. Also
    a horn design worked much better than expected.
Ask a Question
Want to reply to this thread or ask your own question?
You'll need to choose a username for the site, which only take a couple of moments (here). After that, you can post your question and our members will help you out.
Electronics Point Logo
Continue to site
Quote of the day