2025-12-01

2026-03-17 14:58:51 -06:00
parent 183e865f8b
commit 4b82b57113
6846 changed files with 954887 additions and 162606 deletions
@@ -0,0 +1,30 @@
+;; Copyright 2015-2021 Mathieu Bernard
+;;
+;; This file is part of phonemizer: you can redistribute it and/or
+;; modify it under the terms of the GNU General Public License as
+;; published by the Free Software Foundation, either version 3 of the
+;; License, or (at your option) any later version.
+;;
+;; Phonemizer is distributed in the hope that it will be useful, but
+;; WITHOUT ANY WARRANTY; without even the implied warranty of
+;; MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+;; General Public License for more details.
+;;
+;; You should have received a copy of the GNU General Public License
+;; along with phonemizer. If not, see <http://www.gnu.org/licenses/>.
+
+;; This script is executed by festival for English text phonemization.
+(define (phonemize line)
+  "(phonemize LINE)
+Extract the phonemes of the string LINE as a tree and write it to stdout."
+  (set! utterance (eval (list 'Utterance 'Text line)))
+  (utt.synth utterance)
+  ;; Use of print instead of pprintf to have each utterance on one line
+  (print (utt.relation_tree utterance "SylStructure")))
+
+;; This double braket have to be replaced by the name of the text file
+;; you want to read data from. To be parsed by festival as a unique
+;; utterance, each line of that file must begin and end with
+;; double-quotes.
+(set! lines (load "{}" t))
+(mapcar (lambda (line) (phonemize line)) lines)
@@ -0,0 +1,27 @@
+a ʌ
+â aː
+b b
+ch tʃ
+d d
+e eː
+f f
+g g
+h h
+i ɪ
+î iː
+j dʒ
+k k
+kw kʷ
+l l
+m m
+n n
+o ʊ
+p p
+s s
+sh ʃ
+t t
+th θ
+u ʊ
+û o
+w w
+y j
@@ -0,0 +1,27 @@
+a	ʌ
+â	aː
+b	b
+ch	tʃ
+d	d
+e	eː
+f	f
+g	g
+h	h
+i	ɪ
+î	iː
+j	dʒ
+k	k
+kw	kʷ
+l	l
+m	m
+n	n
+o	ʊ
+p	p
+s	s
+sh	ʃ
+t	t
+th	θ
+u	ʊ
+û	o
+w	w
+y	j
@@ -0,0 +1,20 @@
+a	a
+g	g
+h	h
+i	i
+j	j
+k	k
+l	l
+ll	ɬ
+m	m
+n	n
+ng	ŋ
+nng	ŋŋ
+p	p
+q	q
+r	ʁ
+rng	ɴ
+s	s
+t	t
+u	u
+v	v
@@ -0,0 +1,35 @@
+a	a
+aa	aː
+b	b
+by	bʲ
+ch	tʃ
+d	d
+e	e
+ee	eː
+f	ɸ
+g	g
+gy	gʲ
+h	h
+hy	ç
+i	i
+j	dʒ
+k	k
+ky	kʲ
+m	m
+my	mʲ
+n	n
+ny	ɲ
+o	o
+oo	oː
+p	p
+py	pʲ
+r	r
+ry	rʲ
+sh	ʃ
+t	t
+ts	t͡s
+u	ɯ
+uu	ɯː
+w	w
+y	j
+z	z
@@ -0,0 +1,38 @@
+a	a
+b	b
+ch	tʃʰ
+d	d
+e	e
+f	f
+g	χ
+h	h
+hl	ɬ
+i	i
+j	dʒ
+k	k
+kg	kx
+kh	kʰ
+l	l
+m	m
+n	n
+ng	ŋ
+nq	ǃ̃
+ny	ɲ
+o	o
+p	t
+ph	pʰ
+q	ǃ
+qh	ǃʰ
+r	r
+s	s
+sh	ʃ
+t	t
+th	tʰ
+tj	tʃ
+tl	tɬ
+tlh	tɬʰ
+ts	t͡s
+tsh	t͡sʰ
+u	u
+w	w
+y	j
@@ -0,0 +1,45 @@
+a	a
+aa	aː
+aʼ	a̰
+aʼa	a̰ː
+b	b
+ch	t̠͡ʃ
+chʼ	t̠͡ʃʼ
+e	e
+ee	eː
+eʼ	ḛ
+eʼe	ḛː
+f	f
+h	h
+i	i
+ii	iː
+iʼ	ḭ
+iʼi	ḭː
+j	x
+k	k
+kʼ	kʼ
+l	l
+m	m
+n	n
+ñ	n
+o	o
+oo	oː
+oʼ	o̰
+oʼo	o̰ː
+p	pʼ
+pʼ	pʼ
+qu	k
+r	r
+s	s
+x	ʃ
+t	t
+ts	t͡s
+tsʼ	t͡sʼ
+tʼ	tʼ
+u	u
+uu	uː
+uʼ	ṵ
+uʼu	ṵː
+w	w
+y	j
+z	s