I would like to go about streaming voice...to my PC speakers.
First, you have to figure out how to get voice quality data on the Arduino. That generally requires sampling the microphone more often than the Arduino is capable of.
Now, what data are you going to get from the microphone? Just volume data, or frequency and volume data?
Now, how are you going to send that data via the ethernet shield? Generally VOIP uses UDP. Do you understand what that means? Do you know how to send UPD packets? Do you know how much data you can put in each packet? Do you have any idea how long it takes to construct and send a packet? Do you have any idea how many packets you need to send in any given period of time?
Now, suppose you figure out the answers to these questions, and that, by some miracle, the project still looks practical. Do you have any clue how to subscribe to those UDP packets, and to collect those UDP packets and get the data from them, on the PC? Do you have any idea how to use that data to make the speaker do something?
any guidance would be GREATLY appreciated.
I'd suggest looking for a different project.