You have specified a particular audio frequency to four significant figures: 146.8 Hz, which happens also to be musical note D3 on the equally tempered chromatic scale used in America (and elsewhere). Is there some reason for selecting this particular frequency?
Please realize that there is likely to be a range of frequencies picked up by your microphone simultaneously, some above and some below 146.8 Hz. What do you want your LEDs to do if this occurs?
What is lacking is hysteresis in your specification of 146.8 Hz. In the real world, almost nothing is exact, but what if the sound frequency picked up by your microphone was exactly 146.8 Hz? Would the red LEDs or the blue LEDs light up? Or would both light up? You mentioned in a later post that the LEDs are to be an effects panel built into the guitar. That probably means there is a lot of latitude in their operation, so long as you achieve the "desired" effect!
If you want exclusivity in the LEDs, then the circuit needs hysteresis. There must be an "error band" of plus and minus frequencies centered on 146.8 Hz. The last sound frequency that was
less than 146.8 Hz, minus the lower "error band" frequency, results in the red LEDs illuminating. The last sound frequency that was
more than 146.8 Hz, plus the higher "error band" frequency, results in the blue LEDs illuminating. If the current sound frequency is within the "error band" then the previous state of LED illumination does not change. This is hysteresis and it is necessary for stable operation of the LED readouts.
Hysteresis can be as small as you like but never zero. Extremely small hysteresis is difficult to obtain, and may not be useful for your purpose. A reasonable hysteresis might be +/- 0.10 Hz. That means any sound frequency greater than 146.90 Hz lights the blue LEDs while any sound frequency lower than 146.70 Hz lights the red LEDs. Any sound frequency between 146.70 Hz and 146.90 Hz results in no change of the previous LED illumination condition.
So, how do you achieve your goal? I think I would use two active filters, one a high-pass and the other a low-pass, with a corner-frequency of 146.8 Hz. From there it can get a lot more complicated. You need some way to compare energy coming from the low-pass filter to energy coming from the high-pass filter and decide which is greater than the other, so as to select the LEDs (red of blue) to light. Since this is a guitar effects display, a simple full-wave rectifier on each filter output could be applied to a comparator to perform the selection, but you would also need a threshold comparator ahead of the filters to blank the LEDs if there are no strings being played.
The roll-off characteristics of the filters could be important, but this is determined by the type of filter and the filter order. See
this Wikipedia article. I would tentatively suggest a second-order Butterworth filter with 12 db per octave roll-off to start with. These are fairly simple to implement with a few op-amps and a handful of resistors and capacitors. I am probably too old to mess around with musical stuff anymore, and it would behoove you to keep this gadget as analog as possible, so I am going to suggest that someone in this forum,
@KrisBlueNZ for example, chime in with a circuit you can try.
