Multilingual support
The table below gives an overview of the multilingual support for each feature area in ELFEN.
Feature Area |
Supported Languages |
Notes |
|---|---|---|
surface |
All languages with available spacy/stanza models |
raw_sequence_length is language-agnostic |
pos |
All languages with available spacy/stanza models |
|
lexical richness |
All languages with available spacy/stanza models |
|
readability |
All languages with available spacy models and spacy_syllables support |
Most readability formulas were designed for longer English texts. |
information |
Language agnostic |
|
semantic |
Open WordNet-based features: All languages with available spacy/stanza models and OWN support; Hedges: English only |
|
entities |
All languages with available spacy/stanza models |
|
emotion |
All languages with available spacy/stanza models and available lexicons. For the full specification of the available, check the original resources. |
|
psycholinguistic |
language dependent, currently EN, DE, FR, IT, ES, PL, NL supported. For more information, see full table |
More languages planned, psycholinguistic features are dependent on available norms that have to be integrated. |
morphological |
All languages with available spacy/stanza models |
|
dependency |
All languages with available spacy/stanza models |
For psycholinguistic norms, the following table provides an overview of the available resources for each language: .. csv-table:
:header-rows: 1
:file: ../psycholinguistic_norms.csv
:widths: 30, 10, 10, 10, 10, 10, 10, 10