Combining voicebanks to use as appends?

[krai]

Since I enjoy using Amaga Elu a lot, I figured I could try to merge her whisper voice with her normal voicebank to switch between them in the same UST. I've tried to use setparam, editing the otos, but no luck.
Her core voicebank is a multipitch and her whisper voicebank is a VCV recorded on a single pitch. Her whisper oto.ini has the whisper suffix (囁) in the alias, so I tried to put the .wav, .frq and oto in her core folder and tried to switch to the whisper bank using suffixbroker, but she made no sound.
I don't know if I should put the suffix somewhere in the filenames, and I still wouldn't know where to actually store the whisper voice's files. Can I just leave them in the core folder or should I make another folder inside the core voice's one? Both voicebanks are VCV by the way.

Spoiler

her whisper oto and her core voicebank's folder, both unedited

 

[Lyouuv]

Oooh I've done this a lot!!! I think you're right about putting her whisper pitch directly into the normal voicebank folder, if that's what you mean.

To use Elu's wisper alongside her normal voicebank at the same time, I'd just move her whisper pitch folder (the folder containing all the .wav and frq files) and put it in her normal voicebank folder, next to all her other normal pitches in the folder. Sometimes I rename the folders of pitches that were originally from a different voicebank so that I know specifically where they're from. Keep the 囁 alias in the whisper folder. Then when you go to use her in UTAU you'll have to add the 囁 suffix at the end of a note to call the whisper pitch.

You won't need to switch to her whisper, because technically you'll be using both at the same time. As long as you keep the 囁 alias in the oto, and have the whisper pitch folder sitting in the root of her normal voicebank folder (alongside all her normal pitches) it will work!

It could look something like this:

<Elu Amaga Normal Voicebank folder>
- <normal pitch>
- <normal pitch>
- <normal pitch>
- <whisper pitch>
- character.txt
- readme.txt
- oto.ini
- etc

 

[Kiyoteru]

The problem you may be encountering is that the lyric you input is something like [- か囁], but UTAU is applying the prefix map and trying to play the sample [- か囁D4], which doesn't exist. Put a question mark at the start of the lyric to disable the prefix map, like this: [?- か囁], and UTAU will try to look for a sample named [- か囁].

 

[krai]

it just doesn't work for some reason. i moved the whisper folder inside the normal folder and even went through the tedious process of renaming the .wav, .frs and editing the oto to add the whisper suffix at the end of every file, but her whisper voicebank just won't work. the normal one works just fine...
the normal voicebank has a prefix map file while the whisper bank doesn't, could it relate to that somehow?

 

[Lyouuv]

it just doesn't work for some reason. i moved the whisper folder inside the normal folder and even went through the tedious process of renaming the .wav, .frs and editing the oto to add the whisper suffix at the end of every file, but her whisper voicebank just won't work. the normal one works just fine...
the normal voicebank has a prefix map file while the whisper bank doesn't, could it relate to that somehow?

Oh that is really strange... and yeah the whisper missing a prefix map is probably the issue then!! That means UTAU wouldn't know when to play the any of the whisper notes no matter the pitch. Honestly if I were you I'd try re-downloading the whisper voicebank because it could be that either something got corrupted when downloading/installing or maybe you touched something to mess up the prefix map, or something?

(Slightly related but a few days ago I had an oto that randomly deleted itself as I was editing it, and I'm wondering if something like this would have happened to your prefix map while you were editing it? I actually made a big post about being stressed about it and asking for a solution, then @Kiyoteru said that it could've just been then that I had UTAU installed inside my Program Files folder and that editing things like oto's/prefix maps can be problematic, and that I should move my UTAU folder outside of Program Files. So maybe try that too! I just dropped mine onto my desktop.)

Anyways so yeah I'd re-download the whisper voicebank (and then check to see if it has a prefix map lol), move the whisper pitch folder straight into the normal voicebank. If the oto comes with the 囁 alias then keep it, and you don't need to rename any .wavs or .frq files. I don't think you ever need to rename .wavs as long as every phoneme is unique in the oto, meaining it has an alias like 囁 or a pitch at the end like G3.

 

[krai]

Yup, I think the issue might just be the missing prefix map! I redownloaded her whisper bank and there is no prefix map anywhere. Should I make a separate prefix map and put it in the whisper folder or can I edit the normal voicebank's map?
My UTAU folder is on my desktop, yup ^^' I already had some issues with the program files directory in the past haha.

 

[Lyouuv]

Should I make a separate prefix map and put it in the whisper folder

I think that's the right way to go about it? uh I've never had to create a prefix map of my own before so I can't really say, sorry;; And maybe you've already done this but did you check her website to make sure that there wasn't another place to download her whisper prefix map separately? It just seems so weird she'd come without one, but anyways.

Try making a prefix map for the whisper voicebank, and after that put the whisper pitch folder with the prefix map in it into the normal voicebank folder. You could also double check the prefix map again once you have merged the two voicebanks just to make sure things look normal? And that the whisper pitch is included somewhere on the prefix map. That's what I'd try at least.

 

[krai]

Good news, the process may actually be simpler than we imagined! I decided to run the whisper vb's readme through a translator and I got a pretty confusing but kinda helpful explanation....

Spoiler

"Replace the "otome.ini" in the same folder layer as this readme.txt with "oto.ini" and replace "Ameuta el extend/whisper" with ameuta el new record If you plunge each folder into a layer with a folder of continuous sounds, an integrated library will be created,
You can use different voice qualities by using surffixbroker in the same ust. It is the same procedure as Tetsu Kasane. (Reference URL: http://kasaneteto.jp/ongen_howto7.html) Please try it if you like.
(There is a similar oto.ini in the other folder, so when doing the above, please change the other folder as well!)"

It's got a weird wording so it's a bit tricky to fully figure out but I guess all it takes is some more trial and error