That bit in the docs about NV not being chromatically sampled refers to that you are restricted melody-wise to what has been sung. However, many phrase elements are repeated by the singers so you may be able to find them in the right note but that's not guaranteed.
EDIT: that's also the reason why NV uses the VTPC script, which is geared towards manipulating and shaping these sampled phrases, while VT (which powers Soprano and Tenor) operates on chromatically sampled material (where you can do build your own melody-lines).
I am sorry if anything written on the product page and the guide has not answered your questions. What you are asking for is achieved via the Tenor and Soprano - not Native Voice.
For the reason that ongoing Vocal Tools Classical Edition (Tenor and Soprano) is chromatically sampled and not phrases in the traditional sense, you have the ability to compose lyrical vocal lines in the key and tempo of your choosing as a multi array of vocal elements (vee chay nah tus, etc) is strung together with a powerful and real-time legato.
Though NV is sung in many keys, it is not a chromatic.
But what can be done with the conventional phrase?
. Juke box style phrase sequencing via multi list drop down menu.
See what you hear! Phrase abbreviations and vocal keys are clearly listed.
. Real-time advanced sample offset for linear phrase manipulation accuracy.
. Real-time mod control of vocal speed/tempo.
. Assignable MIDI CC control of vocal attack, release and more.
. Multi verse creation banks captured via key switches for convenient call back.
. Breath samples with command of random occurrence and mix volume.
. Select IRF, NI reverb or studio dry all within the scripted environment.
If you still have questions, don't hesitate to ask!
What was confusing ME (I don't say that the information provided by Bela D is in any way wrong or misleading) in the ads for Native Voice, is the remark that the CLASSICAL EDITION of Vocal Tools is chromatically sampled. This information is absolutely clear and correct, but made me wonder how Native Voice is sampled. Maybe the information in your reply "Though NV is sung in many keys, it is not a chromatic." would help to avoid any possible confusion.
(BTW Of course, I'm spoilt by the quality of the chromatically sampled Bela D libraries, which offer an enormous flexibility and have raised the bar for sampled vocals to a new height. I can imagine that this kind of flexibility is just not possible (yet) with a native voice library. Judging by the demos, Native Voice does a very fine and convincing job and is absolutely state of the art.)