フリーのアメリカ英語用発音辞書であるCMUdict (The Carnegie Mellon University
Pronouncing Dictionary)をIPA/X-SAMPA表記に変換したものです。UTAU向けの英語音
源開発、及びUTAUで英語の歌を歌わせる際の発音記号への変換等に使われる事を想定
しています。
サンプル:
ACKNOWLEDGE æknˈɑːlɪd͡ʒ { k n A l I dZ
ACKNOWLEDGE(1) ɪknˈɑːlɪd͡ʒ I k n A l I dZ
ACKNOWLEDGEABLE æknˈɑːlɪd͡ʒəbəl { k n A l I dZ @ b @ l
ACKNOWLEDGEABLE(1) ɪknˈɑːlɪd͡ʒəbəl I k n A l I dZ @ b @ l
ACKNOWLEDGED æknˈɑːlɪd͡ʒd { k n A l I dZ d
$ egrep -i '^sing' dict.txt
SING sˈɪŋ s I N
SING'S sˈɪŋz s I N z
SINGAPORE sˈɪŋəpˌɔːɹ s I N @ p O r\
SINGAPORE'S sˈɪŋəpɔːɹz s I N @ p O r\ z
...
・X-SAMPAシーケンスで検索(完全一致)
$ egrep ' oU z$' dict.txt
EAUX(1) ˈoʊz oU z
O'S ˈoʊz oU z
...
OOOHS(1) ˈoʊz oU z
OSE ˈoʊz oU z
OWES ˈoʊz oU z
$ egrep ' k [^ ]+ t$' dict.txt
CAT kˈæt k { t
CATE kˈeɪt k eI t
CATT kˈæt k { t
CAUGHT kˈɑːt k A t
CAUGHT(1) kˈɔːt k O t
・母音→子音のbigram作成
$ for v in i .\?I e { A V O .\?U u @ @\` 3\`; do printf "%8d $v N\n" `egrep " $v N( |$)" dict.txt | wc -l`; done | sort -rn
7538 .?I N
956 { N
357 V N
274 O N
263 A N
254 e N
72 @ N
35 .?U N
20 i N
17 u N
0 @` N
0 3` N
強引に（連続音録音用の）5モーラ区切りで録音できるように
5 syllable x 20に詰め込んでみた。発音できるかはシラネｗ
i I e E {
A V O o U
a u @ @` 3`
a U a I @
I e I e @
o I o U @
p_hi pI pr\e bE br\{ p
t_hA tV tr\O doU dr\o t
k_hi kI kr\e gE gr\{ k
i f@ f3` v@ v3`f
A TV DO r\oU bloT
e s@ S3` z@ Z3` s
dZi tSI tse dzE h{
u maU maI noU noI
I jeI weI lu@ NoU
O dr\u pI@ fe@ gwe b
U klU toI poI r\@` d
{ kwA swV fr\a dw@` g
V Tr\O flo blu k3` v
o gl e twi pl@ Ti N
(一応、歌詞)
It's a joke,it's a joke,it's all lies What I told were all,but fibs
It's a joke,it's a joke,it's all lies You are an idiot!Hey it's a joke.
Elvis is alive! It's a joke.MJ is alive! It's a joke.
T.Rex is back,Blues Brothers is back,Beatles is... It's a joke.
原曲は「重音テトの嘘八百」です。
