Wow, we need some basic basics here.

Will WAV or WMA "work" for karaoke or not? WAV, WMA, MP3, M4A.... They are just music files like every other music file. They will all "work". I don't know what you mean by "work". That's like asking if the "Ford gas" is okay for your Chevy. The gas doesn't know what car it's going into any more than that file knows what it is being played back on, or where. Don't define karaoke as anything more than you playing songs with no vocal track just like you play any other song.

That out of the way....

This topic seems to have shifted direction a few times. Yes, you need a codec (code-decode is like "modem" is modulate-demodulate) to encode a data file into a specific format. That "specific format" is a compressed format, the value add of which is mainly that compressed MP3 files are like 10% the size of the uncompressed WAV file. WAV files sound a little brighter and cleaner than MP3 because they have not been compressed.

Now factor in at what rate you want to compress the WAV files to MP3. 128? 220? 260? 320? The encoders that come on your computer and with BIAB/RB will not go up to 320. (RHARV, they go to what, 160 is it?)

Thus, many of us opt to use another piece of software to be able to convert those WAV files to MP3 files, again, largely for space considerations and transmission by email. Audacity is one. Adobe Audition is another. I opt for Audition just because it's what I know. You will find people here defend their chosen software like they protect their children from alien attack. It all comes down to what you are comfortable.

Someone mentioned SONAR. There is a downloadable plugin that works very well IF you have a legal, purchased, licensed copy of SONAR. On the occasions that I am mixing in tracks in SONAR, I use their built-in codec to compress to MP3. If I am satisfied with the mixing in RB, I will use Audition. And the reason I would use one or the other is this.

I have a control surface that records mixer moves via automation, and as I mix down I do a lot of fading, muting, etc, things that are probably available in RB but I don't know how to do them, and SONAR remembers them and moves the mixer faders during playback. Like if I want to mute a rhythm guitar track from bar 19 through bar 25, in SONAR I can swipe it with the mouse, mute that section, listen to it, and keep or undo if I like it or not.

I really think you are WAY overthinking the role of the codec and such. Just use the menu pulldowns and it will walk you right through the process. For karaoke, an MP3 compressed to 128 will be fine.