nb_g2p 2.1.1

Transcribe Norwegian dialects phonemically from bokmål text.

Grapheme to Phoneme models for Norwegian Bokmål

This repo contains code to run G2P models for Norwegian bokmål[^1], which produce phonemic transcriptions for close-to-spoken pronunciations (such as in spontaneous conversations: spoken) and close-to-written pronunciations (such as when reading text aloud: written) for 5 different dialect areas:

1. East Norwegian (e)
2. South West Norwegian (sw)
3. West Norwegian (w)
4. Central Norwegian (Trøndersk) (t)
5. North Norwegian (n)

[^1]: Bokmål is the most widely used written standard for Norwegian. The other written standard is Nynorsk. Read more on Wikipedia.

Setup

pip install nb_g2p

Usage

>>> import nb_g2p
>>> list(nb_g2p.transcribe("hei på deg!"))
[('hei', 'H AEJ1'), ('på', 'P OAH0'), ('deg', 'D AEJ1')]

Transcription standard

The G2P models have been trained on the NoFAbet transcription standard which is easier to read by humans than X-SAMPA. NoFAbet is in part based on 2-letter ARPAbet and is made by Nate Young for the National Library of Norway in connection with the development of NoFA, a forced aligner for Norwegian. The equivalence table below contains X-SAMPA, IPA and NoFAbet notatations.

X-SAMPA-IPA-NoFAbet equivalence table

X-SAMPA	IPA	NoFAbet	Example
A:	ɑː	AA0	bad
{:	æː	AE0	vær
{	æ	AEH0	vært
{*I	æɪ	AEJ0	sei
E*u0	æʉ	AEW0	sau
A	ɑ	AH0	hatt
A*I	ɑɪ	AJ0	kai
@	ə	AX0	behage
b	b	B	bil
d	d	D	dag
e:	eː	EE0	lek
E	ɛ	EH0	penn
f	f	F	fin
g	g	G	gul
h	h	H	hes
I	ɪ	IH0	sitt
i:	iː	II0	vin
j	j	J	ja
k	k	K	kost
C	ç	KJ	kino
l	l	L	land
l=	l̩	LX0
m	m	M	man
n	n	N	nord
N	ŋ	NG	eng
n=	n̩	NX0
o:	oː	OA0	rå
O	ɔ	OAH0	gått
2:	øː	OE0	løk
9	œ	OEH0	høst
9*Y	œy	OEJ0	køye
U	u	OH0	f*ort
O*Y	ɔy	OJ0	konvoy
u:	uː	OO0	bod
@U	oʉ	OU0	show
p	p	P	pil
r	r	R	rose
d`	ɖ	RD	rekord
l`	ɭ	RL	perle
l`=	ɭ̩	RLX0
n`	ɳ	RN	barn
n`=	ɳ̩	RNX0
s`	ʂ	RS	pers
t`	ʈ	RT	stort
s	s	S	sil
S	ʃ	SJ	sju
t	t	T	tid
u0	ʉ	UH0	russ
}:	ʉː	UU0	hus
v	ʋ	V	vase
w	w	W	Washington
Y	y	YH0	nytt
y:	yː	YY0	ny

Unstressed syllables are marked with a 0 after the vowel or syllabic consonant. The nucleus is marked with a 1 for tone 1 and a 2 for tone 2. Secondary stress is marked with 3.

License

These models are shared with a Creative_Commons-ZERO (CC-ZERO) license, and so are the lexica they are trained on. The models can be used for any purpose, as long as it is compliant with Phonetisaurus' license.