Looking To Make A Comprehensive English Voicebank! Help!

Hi, My name is Sky, I’ve been using Deepvocal to voice a robot character for about a month now.
I’ve noticed there are very few english VBs available. All of them are better at speaking Japanese than English.

I only speak English and would like to help create a voice bank that has ALL of the english syllables.
Ideally this voicebank would use a dictionary of Latin characters.

I have the linguistic side of this mapped out and designed.
I do not know how to actually make a voicebank. I have tried. I watched many tutorials and I still can’t seem to map a single syllable.

I am a professional artist and a pretty okay singer. I can sing in character and have a pretty decent range. I can and will provide the voice clips and the silly character art. I just need someone to help me actually turn my recordings of all the different syllables into a usable voicebank.

Here’s the syllable list:
the vowels are in all capital letters.
the consonants are in all lowercase.
Comic sans used to distinguish lowercase letter L from capital letter I

Each Vowel and consonant would also have its own independent recordings.

The consonants are

b as in boy
d as in dog
f as in frank
g as in gold
h as in hat
ch as in chuck
sh as in shut
j as in jam
k as in kite
sk as in ski
l as in luck
sl as in slug
m as in mom
sm as in small
n as in now
sn as in snap
p as in puck
sp as in speed
q as in queen (q is always a qu sound)
sq as in squeak
r as in rat
s as in side
t as in tank
st as in stick
v as in vet
w as in wide
sw as in swim
x as an end sound like in fox of box
z as in zero

the vowels are a bit more japanese in spelling because the japanese spellings make more sense than the english silent E. The vowels should be in all caps as to not accidentally spell two sylables the exact same way.

A is said ah as in sash
E is said eh as in best
I is said ee as in see
O is said oh like the oa in goat or boat
U is said oo like in hoop
AY is said A as in name
AI is said like the word “I”
IH makes the traditional lowercase i sound like inside
AW is the traditional lowercase o sound like open
YU is said you
UH is the traditional sound for lowercase u like in sun

because it’s so comprehensive syllable wise it would likely work in other languages as well.

if someone can even simply turn this into an operating dictionary that I could just copy paste, I could do most of it myself. I definitely wont turn down help.


heres a rough sketch of the little DVloid. I think I’ve settled on calling her dIHtzI (ditzy)
please let me know if you can help.


What you’ve got here is a classic CV-VC reclist. Easy to program in both UTAU and DV. However, I see a lot of missing consonant clusters. Is this all you have planned for your reclist? Or is this just a sample?

Either way, I’d be happy to lend a hand! I’m working on my own English reclist, so maybe we can learn from each other?

I began working on an English VB. If you want it to be comprehensive, you will need a lot of samples. I made a reclist and it is quite huge (I already have over 7000 samples and I still miss some final CC, VV transitions and some common CC transitions) but covers a lot of sound combinations. If you need help, I can give some advice