Dataset Viewer
Auto-converted to Parquet Duplicate
language
stringclasses
1 value
source
stringclasses
1 value
concept_id
int64
1
100
english
stringlengths
1
6
asjp
stringlengths
2
13
wob
ASJP
1
I
mo*
wob
ASJP
2
you
mo, amo*
wob
ASJP
3
we
apuo
wob
ASJP
11
one
to
wob
ASJP
12
two
so*
wob
ASJP
19
fish
smi
wob
ASJP
21
dog
gb~e
wob
ASJP
23
tree
tu
wob
ASJP
28
skin
ku
wob
ASJP
30
blood
nm~o
wob
ASJP
31
bone
kp~a
wob
ASJP
39
ear
do*u*
wob
ASJP
40
eye
Tria
wob
ASJP
41
nose
mla
wob
ASJP
43
tooth
5n~ea
wob
ASJP
44
tongue
meo
wob
ASJP
47
knee
kou
wob
ASJP
48
hand
kw~a
wob
ASJP
54
drink
na
wob
ASJP
57
see
Te
wob
ASJP
58
hear
Nw~o
wob
ASJP
61
die
ma
wob
ASJP
66
come
Ti
wob
ASJP
72
sun
Tru
wob
ASJP
75
water
ni
wob
ASJP
77
stone
sou
wob
ASJP
82
fire
na
wob
ASJP
85
path
sru
wob
ASJP
92
night
ma*o*, kw~iye
wob
ASJP
100
name
5n~e

Open Wobé lexicon leads

ASJP Wobé wordlist

Field Value
File data.jsonl (row-based dataset)
Source https://asjp.clld.org/languages/WOBE
Language Wobé / Wè Northern (wob, weno1238)
Entries 30 concept-word rows in data.jsonl
Source publication Hofer and Link (1980), West African language data sheets, vol. 2
License CC BY 4.0 for the ASJP database
Representation ASJP phonological transcription, not the current Wobé practical orthography
MT value Tiny lexicon / normalization and terminology-check seed; not parallel-sentence training data

The JSONL file is provided so the Hugging Face viewer and datasets.load_dataset() expose one row per ASJP concept. The original downloaded ASJP record is retained in the project repository for provenance. The ASJP forms are phonological transcriptions, not verified practical-orthography Wobé spellings.

The English concept labels can be translated to French for a small bilingual evaluation glossary, but any conversion from ASJP transcription to the project orthography must be reviewed by a Wobé linguist. Do not silently merge these forms into the Bible text.

Downloads last month
53