Pulse to the Beat

For the past 3 weeks I've been working on this, and now I'm stuck. I need some help, so here it goes.

I'd like to design a circuit that will allow me to pulse EL wire to a beat. It's for an EL Wire Sweatshirt. I have the EL Wire controller circuits working great. My goal is to design a circuit I can feed an Arduino Fio, powered by a 3.7VDC (850mAh) LiPo, that listens to music from a microphone, amplifies it, finds the beat, then feeds the ADC. I've tried 6 different methods and I'm ready to give up. How can I do this meeting my power requirements, the Fio's limited ADC range, while using a microphone (not a 1/8" jack!).

I am proficient in coding for the Arduino, and I have all the other circuitry working. I just need a means of making collecting and analyzing the audio data so I can use it with the Audio circuit. I have no problem buying parts if needed.

I will appreciate any help I can get!

Curiosity eats me up, what are 6 methods ? I know only one, filtering, which could be done using hardware or software. May be second methods, w/o filtering, just run AC measurements, as beat represents low part of spectrum, the most energy powerfull, it will drive all output up and down. It'd left 4 mystery

I looked at 2 different Amp circuits that ran 5 V (didn’t know they’d not work).
Used a 3.3v circuit, too much noise next to the AC stuff.
Tried a BJT CE Amp, not enough amplification.
Tried a RadioShack kit, it provides too much amplification, it was a triple NPN transistor circuit. Range was 0-7.5V. Too large and it had a separate battery.

So it was 5, not 6 my bad.

LM358 OPA should be o'k down to 3V according to data sheet: "• Wide power supply range: Single Supply: 3V to 36 V" Schematic for mic pick-up you can find here: http://coolarduino.wordpress.com/2012/06/22/audio-input-to-arduino/ I used NE5532, but not sure it will works with less than +5V.

Would you mind testing this for me? I’ve tried these circuits saying the amp can run on 3V but it misbehaved. The microphone and amp need to be powered by 3.7V, not just the amp. Will it still work in these circumstances?

I can't see why it should not run on 3.7V. OPA LM358 warranted by manufacturer to operate at 3V. Regular electret mic is good down to 2V, it says specification report: "Test Condition (Vs= 2.0 V, RL= 2.2 k ohm, Ta=20°C, RH=65%)"... Link: https://www.sparkfun.com/products/8635?

I've been prototyping for weeks with these circuits, hence I'm wary, but if it's rated, may as well try...