AMPR GA Correspondences for Ponto

AMPR is a scoring system designed for young writers of rhotic varieties of English, especially General American. It is a non-exhaustive parsing scheme, meant to be run with the Insertion penalty to 0. The stimuli are allophonic IPA transcriptions, as discussed in more detail below.

stimulus	response	penalty
t̠	t	0
t̠	c	0
d̠	d	0
d̠	g	0
d̠	j	0
b̥	b	0
b̥	p	0
d̥	d	0
d̥	t	0
ɡ̥	g	0
ɡ̥	c	0
ɡ̥	k	0
ɡ̥	q	0
g̥	g	0
g̥	c	0
g̥	k	0
g̥	q	0
d̥̠	c	0
d̥̠	d	0
d̥̠	g	0
d̥̠	j	0
tʃ	c	0
dʒ	g	0
dʒ	j	0
aɪ	i	0
aɪ	y	0
aʊ	o	0
ɔɪ	o	0
θ	t	0
ð	t	0
ʃ	s	0
ʒ	s	0
r	r	0
ɡ	g	0
g	g	0
ŋ	n	0
ɪ	i	0
ɛ	e	0
ɛ	a	0
æ	a	0
ɚ	r	0
ɝ	r	0
ə		0
ʊ	u	0
ʊ	o	0
ʌ	u	0
ɔ	o	0
ɔ	a	0
ɑ	o	0
ɑ	a	0
p	p	0
b	b	0
m	m	0
f	f	0
v	v	0
w	w	0
w	y	0
t	t	0
d	d	0
n	n	0
k	c	0
k	k	0
k	q	0
h	h	0
j	y	0
s	s	0
s	c	0
z	z	0
z	s	0
l	l	0
i	e	0
i	y	0
e	a	0
u	u	0
u	o	0
o	o	0
ɾ	t	0
ɾ	d	0

Documentation

This implements the metric used in Treiman & Kessler (2004) Applied Psycholinguistics, 25, 413–428. The correspondence set looks for single letters as spellings for allophones. It is designed to be used with the checkbox “Require correct sequence” turned off, so that extra letters in the spelling, including legal elements in digraphs, are ignored without penalty. In fact, even sequential order is ignored, so that “attack” would be considered a fully correct spelling for “cat”. Obviously this is intended for use with very young children. Transcription is partly allophonic, making transcription somewhat more complicated than for most other methods. The AMPR IPA version requires stimuli to be transcribed in IPA; AMPR SAMPA uses X-SAMPA conventions, which are easier to type but harder to read.

AMPR scoring operates to a certain extent on the allophonic level. The following transcriptional conventions are assumed:

b̥ etc. are lenis voiceless stops, as found after /s/, equated with either /b/, etc., or /p/, etc.
t̠, d̠ are postalveolar noncontinuants, found before /r/, variously equated with either /t/, /d/ or with /tʃ/, /dʒ/
[d̥̠] lenis AND retracted (as /t/ in /str/ street)
Flap [ɾ] (/t/ or /d/ after a vowel and before an unstressed vowel)

The following table illustrates the phonetic transcription system used in the AMPR method:

Phone	Rule	Example
æ	add	æd
aɪ	ice, spy	aɪs sb̥aɪ
aʊ	out	aʊt
b	bit	bɪt
b̥	b or p after s: spy	sb̥aɪ
d̠	d before r: dream	d̠rim
d	do	du
d̥̠	t between s and r: stream	sd̠̥rim
ð	then	ðɛn
d̥	t or d after s or f, but not before r: sty, after	sd̥aɪ æfd̥ɚ
dʒ	jump, gem, cage	dʒʌmp dʒɛm kedʒ
ə	about, walking (use for any reduced unstressed vowel)	əbaʊt wɔkəŋ
e	ape	ep
ɛ	ebb	ɛb
f	foe	fo
h	hot	hɑt
i	eat, pretty	it prɪɾi
j	yes	jɛs
k	cod, keep, quick	kɑd kip kwɪk
l	love	lʌv
m	me	mi
ŋ	sing	sɪŋ
n	not	nɑt
ɔ	caught (or use ɑ)	kɔt kɑt
o	open	opən
ɔɪ	oyster, noise	ɔɪsd̥ɚ nɔɪz
p	pit	pɪt
r	read	rid
s	see, cent	si sent
ɝ	bird (OK to use ɚ)	bɝd bɚd
ʊ	book	bʊk
ɚ	butter	bʌɾɚ
ɡ̥	c /k/, k, q after s: ski, squid	sɡ̥i sɡ̥wɪd
ɑ	cot (or use ɔ)	kɑt kɔt
ɡ	go	ɡo
ɪ	id	ɪd
ʃ	she	ʃi
ɾ	t or d between vowel and unstressed vowel: latter, ladder	læɾɚ læɾɚ
ʌ	ugly (stressed)	ʌɡli
tʃ	chip	tʃɪp
t̠	t before r: truck	t̠rʌk
t	toy	tɔɪ
u	ooze	uz
v	vie	vaɪ
w	wide	waɪd
z	zoo, is	zu ɪz
ʒ	pleasure	plɛʒɚ
θ	thin	θɪn

See also ampr-sampa for a version that uses ASCII substitutes for IPA characters.