I'm going to take a stab at what I think you're getting confused over.
Microphones are monophonic devices. For that matter, so are speakers - you have to pair up multiples of them if you want more channels. In the case of speakers, you often pair up several just to get one channel by combining several with differing efficient frequency response ranges, but that's another story.
Say you want to do voice chat with someone over a computer. What happens is something like this.
- Sound waves impact microphone
- Transducer in microphone creates an analogous electrical signal (same frequency spectrum)
- Electrical signal is conducted to sound card
- Sound card performs analog to digital conversion
- Digital signal goes to VC software
- VC software on your computer sends data to VC software on another computer
- VC software on other end sends data to sound card
- Sound card on other end performs digital to analog conversion
- Analog signal is amplified and signal is sent to output device (speakers, headphones)
- Amplified analog electrical signal causes transducer in other guy's headphones to vibrate, creating sound waves
Okay, the big point is that it is rare today (unless the result is being immediately amplified for an audience) to have microphone data not end up as a bunch of 1s and 0s at some point. At that point, you have a single-channel audio signal. But which channel is it? The left? The right? What if it's going on some 7.1 surround array? Well, it doesn't matter - it's single channel. The convention is that, since hearing voice in only one ear tends to freak humans out slightly, software which is implicitly transmitting voice (e.g. Skype) will assume that you've got a voice transmission coming over your microphone line and automatically clone it to both the Left and Right channels. Other things which don't know what the "meaning" of the content is will generally leave it alone as a single-channel recording / data stream. For playback, it HAS to be sent to at least one physical channel (amplification + speaker) so the convention is to use the Left speaker - which is only "Left" on the basis of where you plugged it in - or if headphones because someone permanently "plugged it in" by wiring it that way. There's no reason it has to be this way, but it's nice when things are consistent because you know what to expect. Sometimes systems with an output channel designated Center will use that instead for monophonic signals - there's not really a right and wrong answer (besides not sending it exclusively to a Subwoofer channel since it will probably not have adequate frequency response).
tl;dr - You don't have to do anything, it almost always works automatically by software if the software is expecting the microphone signal to convey speech.