The amazing Murre (genitive Murren đ) will normalize non-standard Finnish (puhekieli) to standard Finnish (kirjakieli). This repository is maintained by Mika HĂ€mĂ€lĂ€inen.
This library is designed for Python 3 and it may not work on Python 2.
pip3 install murre
python3 -m murre.download
To normalize Finnish, all you need to do is to run:
from murre import normalize_sentence
normalize_sentence("mÀ syön paljo karkkii")
>> minÀ syön paljon karkkia
You can normalize multiple sentences at the same time by running
from murre import normalize_sentences
sents = ["kissa syö karkkii", "jok laulaa tuol puole", "en tiiÀ oikee et kuka se o", "kyl on hölömöö"]
normalize_sentences(sents)
>> ['kissa syö karkkia', 'joka laulaa tuolla puolen', 'en tiedÀ oikein ettÀ kuka se on', 'kyllÀ on hölmöÀ']
To normalize (and lemmatize) historical Finnish, run:
from murre import normalize_sentence
normalize_sentence("paluellen herra caiken", language="fin_hist")
>> palvella herra kaikki
You can use the Swedish model by passing language=swe
from murre import normalize_sentence
normalize_sentence("int vet ja", language="swe")
>> inte vet jag
Murre can also generate different dialects. All you need to do, is to run:
from murre import dialectalize_sentence
dialectalize_sentence("kodin takana on koira", "Inkerinsuomalaismurteet")
>> 'kojin takan on koira'
Or for multiple sentences:
from murre import dialectalize_sentences
sents = ["kissa syö karkkia", "kÀdellÀ on perhonen", "kettu juoksee sutta karkuun"]
dialectalize_sentences(sents,'Kainuu')
>> ['kissa syöpi karkkia', 'kÀellÀ om perhonej', 'kettu juoksee sutta karkuu']
The list of available dialects can be obtained by:
from murre import supported_dialects
supported_dialects()
>> ['Pohjois-Satakunta', 'Keski-Karjala', 'Kainuu', 'EtelĂ€-Pohjanmaa', 'EtelĂ€-Satakunta', 'Pohjois-Savo', 'Pohjois-Karjala', 'Keski-Pohjanmaa', 'Kaakkois-HĂ€me', 'PohjoinenKeski-Suomi', 'Pohjois-Pohjanmaa', 'PohjoinenVarsinais-Suomi', 'EtelĂ€-Karjala', 'LĂ€nsi-Uusimaa', 'Inkerinsuomalaismurteet', 'LĂ€ntinenKeski-Suomi', 'LĂ€nsi-Satakunta', 'EtelĂ€-Savo', 'LĂ€nsipohja', 'Pohjois-HĂ€me', 'EtelĂ€inenKeski-Suomi', 'EtelaÌ-HaÌme', 'PerĂ€pohjola']
Normalization (Finnish)
Niko Partanen, Mika HÀmÀlÀinen, and Khalid Alnajjar. (2019). Dialect Text Normalization to Normative Standard Finnish. In the Proceedings of the 5th Workshop on Noisy User-generated Text (W-NUT).
Normalization (Swedish)
Mika HÀmÀlÀinen, Niko Partanen and Khalid Alnajjar. (2020). Normalization of Different Swedish Dialects Spoken in Finland. In the Proceedings of the 4th ACM SIGSPATIAL Workshop on Geospatial Humanities.
Dialect generation
Mika HÀmÀlÀinen, Niko Partanen, Khalid Alnajjar, Jack Rueter & Thierry Poibeau (2020). Automatic Dialect Adaptation in Finnish and its Effect on Perceived Creativity. In Proceedings of the 11th International Conference on Computational Creativity. p. 204-211
Historical Finnish
Mika HÀmÀlÀinen, Niko Partanen and Khalid Alnajjar. (2021). Lemmatization of Historical Old Literary Finnish Texts in Modern Orthography. In Actes de la Conférence sur le Traitement Automatique des Langues Naturelles (TALN).
The data used in the paper describing dialect generation has been published on Zenodo .