I have yet to see a voice changer that does not leave any form of detectable artifact in the resulting track. If you take it with a grain of salt, of course, it 'sounds like' something. 
But you are left with that metallic, chorus timbre of a processed sound. It will not paint your project in a good light.

As a hobbyist, the best I've seen (and perhaps the most well-known) is the AV Voice Changer Diamond. I hope there is a better or more underground solution for studio professionals.

If you have the money to hire talent you can try a service like Voice123. 
Best of luck! 

Well, if that's the case then you could do all the voice work indeed.
All you need is some help editing the material.
Everytime you want to 'make something sound like other something' you need to start thinking analytically. Let's apply this for a walkie-talkie kind of sound. What gives a sound that walkie-talkie feeling?

Distortion.

By thinking analytically, we can look at a walkie-talkie and see why it produces that characteristic, distorted sound: its speaker has limits; it's often low-quality. I don't have the exact electronic reason why this happens, but I remember it has to do with the transistors in the circuit having a limit on what volume they can reproduce.
With this in mind we can start to sweeten a voice material to reproduce these characteristics. I'd prefer to use SoundForge but since you're using Audacity, let's stick with that.
Start with the original material.
I recorded my own voice and converted it to a Mono track. Used the Compressor filter to put a little more "meat" in the sound. Used the Change Pitch filter to lower my voice so it sounds older:
[attachment=2368:voice1.wav]

Next, I used the Low-Pass and High-Pass filters with subtle settings just to trim the extreme frequencies. This effectively simulates a bad speaker in the sense that it doesn't correctly reproduce very low or very high frequencies. This is similar to a Band-Pass filter (centered on the middle frequencies), which only allows a certain range of frequencies to pass. SoundForge even has a name for this centered Band-Pass preset: "phone-line".
The amount of low frequencies I took was less than the higher ones, because I didn't want the sound to lose too much of its "deepness".
Here is the result:
[attachment=2369:voice2.wav]

Next, we apply the final filter and the one which most gives that characteristic distortion, the Hard Limiter filter.
You'll see that the waveform turns straight at its edges when you apply it - it's not sinewy like it was before. This is the effect of a "hard limiter" or "distortion" filter, limiting a range of samples to a certain volume:
[attachment=2367:voice3.wav]

"How will I know how much I should configure each filter?" use your ears. They're the best judge you have, and you should know that everything you hear the players most likely will too (as long as their speaker isn't too different from the one you're using to monitor the process!). Don't be afraid to go trying out things; you only have to learn.