Board Thread:General Discussion/@comment-72.66.32.27-20141028214010/@comment-25618373-20141202011957

Robyn IsANinja wrote:

EmbraceEvil wrote:

V3-CHAN wrote: I kinda wanna see if "ENGRISH" is any btter for JAP VBs this time around, with the added triphones and such. Doubt it. Engrish seems to depend on the producer (note that I am not saying that engrish is ever a substitute for actual english). Engrish depends on a couple of different factors. One is the producer, yes. If the producer has no idea how to make to align the Japanese phonetics into the closest possible English equivalent, then it sounds strange. Another factor are the voice banks themselves. I only have five Japanese voice banks but they all function a little differently with Engrish. KAITO's soft append, for example, will pronounce the extra phonetics added to a word. Like if you wanted him to say hat, and not hato, you would input [h a t] into the note and he will say it. However, his whisper append fails to do this and requires extra work to make him pronounce the extra t. Also, adding phonetics at the beginning of a note can get complicated and go unpronounced. Like "ski." Most all of KAITO's JP voice banks won't pronounce the s in "ski."

Anon and Kanon are somewhat similar. Kanon has an easier time pronouncing extra phonetics at the beginning of a word while Anon can pronounce any added phonetics at all. I've heard that the same can be said for a lot of newer Vocaloids like Rana and Flower.

So, yes. It would be interesting to see if the JP voice banks will be able to produce better Engrish. I mean, it isn't even just an Engrish problem for me, but rather just a general pronunciation problem. If I want Kanon to say Masu without the U, I have to edit it out myself later in Audacity, but it works fine with KAITO's soft append. (Or perhaps I just favor Kaito too much! Lol) k