Larth: Dataset and Machine Translation for Etruscan

G. Vico; Gerasimos Spanakis

doi:10.26615/978-954-452-087-8_005

Larth: Dataset and Machine Translation for Etruscan

Research output: Chapter in Book/Report/Conference proceeding › Conference article in proceeding › Academic › peer-review

10 Downloads (Pure)

Abstract

Etruscan is an ancient language spoken in Italy from the 7thcentury BC to the 1stcentury AD. There are no native speakers of the language at the present day, and its resources are scarce, as there exist only around 12,000 known inscriptions. To the best of our knowledge, there are no publicly available Etruscan corpora for natural language processing. Therefore, we propose a dataset for machine translation from Etruscan to English, which contains 2891 translated examples from existing academic sources. Some examples are extracted manually, while others are acquired in an automatic way. Along with the dataset, we benchmark different machine translation models observing that it is possible to achieve a BLEU score of 10.1 with a small transformer model. Releasing the dataset1 can help enable future research on this language, similar languages or other languages with scarce resources.

Original language	English
Title of host publication	Proceedings of the Ancient Language Processing Workshop associated with RANLP-2023
Editors	Adam Anderson, Shai Gordin, Bin Li, Yudong Liu, Marco C. Passarotti
Publisher	Association for Computational Linguistics (ACL)
Pages	39-48
Number of pages	10
ISBN (Electronic)	9789544520878
DOIs	https://doi.org/10.26615/978-954-452-087-8_005
Publication status	Published - 8 Sept 2023
Event	ALP Workshop 2023 : 1st Workshop on Ancient Language Processing at RANLP 2023 - Vama, Bulgaria Duration: 8 Sept 2023 → 8 Sept 2023 http://www.wikicfp.com/cfp/servlet/event.showcfp?eventid=172851&copyownerid=180639

Workshop

Workshop	ALP Workshop 2023 : 1st Workshop on Ancient Language Processing at RANLP 2023
Abbreviated title	RANLP 2023
Country/Territory	Bulgaria
City	Vama
Period	8/09/23 → 8/09/23
Internet address	http://www.wikicfp.com/cfp/servlet/event.showcfp?eventid=172851&copyownerid=180639

Access to Document

10.26615/978-954-452-087-8_005

Full TextFinal published version, 462 KBLicence: Taverne

Cite this

@inproceedings{766a50598b8c4724b55191fbe12ace81,

title = "Larth: Dataset and Machine Translation for Etruscan",

abstract = "Etruscan is an ancient language spoken in Italy from the 7thcentury BC to the 1stcentury AD. There are no native speakers of the language at the present day, and its resources are scarce, as there exist only around 12,000 known inscriptions. To the best of our knowledge, there are no publicly available Etruscan corpora for natural language processing. Therefore, we propose a dataset for machine translation from Etruscan to English, which contains 2891 translated examples from existing academic sources. Some examples are extracted manually, while others are acquired in an automatic way. Along with the dataset, we benchmark different machine translation models observing that it is possible to achieve a BLEU score of 10.1 with a small transformer model. Releasing the dataset1 can help enable future research on this language, similar languages or other languages with scarce resources.",

author = "G. Vico and Gerasimos Spanakis",

note = "Publisher Copyright: {\textcopyright} RANLP-ALP 2023 - Proceedings of the Ancient Language Processing Workshop, associated with 14th International Conference on Recent Advances in Natural Language Processing.; ALP Workshop 2023 : 1st Workshop on Ancient Language Processing at RANLP 2023, RANLP 2023 ; Conference date: 08-09-2023 Through 08-09-2023",

year = "2023",

month = sep,

day = "8",

doi = "10.26615/978-954-452-087-8_005",

language = "English",

pages = "39--48",

editor = "Adam Anderson and Shai Gordin and Bin Li and Yudong Liu and Passarotti, {Marco C.}",

booktitle = "Proceedings of the Ancient Language Processing Workshop associated with RANLP-2023",

publisher = "Association for Computational Linguistics (ACL)",

address = "United States",

url = "http://www.wikicfp.com/cfp/servlet/event.showcfp?eventid=172851&copyownerid=180639",

}

Vico, G & Spanakis, G 2023, Larth: Dataset and Machine Translation for Etruscan. in A Anderson, S Gordin, B Li, Y Liu & MC Passarotti (eds), Proceedings of the Ancient Language Processing Workshop associated with RANLP-2023. Association for Computational Linguistics (ACL), pp. 39-48, ALP Workshop 2023 : 1st Workshop on Ancient Language Processing at RANLP 2023, Vama, Bulgaria, 8/09/23. https://doi.org/10.26615/978-954-452-087-8_005

Larth: Dataset and Machine Translation for Etruscan. / Vico, G.; Spanakis, Gerasimos.
Proceedings of the Ancient Language Processing Workshop associated with RANLP-2023. ed. / Adam Anderson; Shai Gordin; Bin Li; Yudong Liu; Marco C. Passarotti. Association for Computational Linguistics (ACL), 2023. p. 39-48.

Research output: Chapter in Book/Report/Conference proceeding › Conference article in proceeding › Academic › peer-review

TY - GEN

T1 - Larth: Dataset and Machine Translation for Etruscan

AU - Vico, G.

AU - Spanakis, Gerasimos

N1 - Publisher Copyright: © RANLP-ALP 2023 - Proceedings of the Ancient Language Processing Workshop, associated with 14th International Conference on Recent Advances in Natural Language Processing.

PY - 2023/9/8

Y1 - 2023/9/8

N2 - Etruscan is an ancient language spoken in Italy from the 7thcentury BC to the 1stcentury AD. There are no native speakers of the language at the present day, and its resources are scarce, as there exist only around 12,000 known inscriptions. To the best of our knowledge, there are no publicly available Etruscan corpora for natural language processing. Therefore, we propose a dataset for machine translation from Etruscan to English, which contains 2891 translated examples from existing academic sources. Some examples are extracted manually, while others are acquired in an automatic way. Along with the dataset, we benchmark different machine translation models observing that it is possible to achieve a BLEU score of 10.1 with a small transformer model. Releasing the dataset1 can help enable future research on this language, similar languages or other languages with scarce resources.

AB - Etruscan is an ancient language spoken in Italy from the 7thcentury BC to the 1stcentury AD. There are no native speakers of the language at the present day, and its resources are scarce, as there exist only around 12,000 known inscriptions. To the best of our knowledge, there are no publicly available Etruscan corpora for natural language processing. Therefore, we propose a dataset for machine translation from Etruscan to English, which contains 2891 translated examples from existing academic sources. Some examples are extracted manually, while others are acquired in an automatic way. Along with the dataset, we benchmark different machine translation models observing that it is possible to achieve a BLEU score of 10.1 with a small transformer model. Releasing the dataset1 can help enable future research on this language, similar languages or other languages with scarce resources.

U2 - 10.26615/978-954-452-087-8_005

DO - 10.26615/978-954-452-087-8_005

M3 - Conference article in proceeding

SP - 39

EP - 48

BT - Proceedings of the Ancient Language Processing Workshop associated with RANLP-2023

A2 - Anderson, Adam

A2 - Gordin, Shai

A2 - Li, Bin

A2 - Liu, Yudong

A2 - Passarotti, Marco C.

PB - Association for Computational Linguistics (ACL)

T2 - ALP Workshop 2023 : 1st Workshop on Ancient Language Processing at RANLP 2023

Y2 - 8 September 2023 through 8 September 2023

ER -

Larth: Dataset and Machine Translation for Etruscan

Abstract

Workshop

Access to Document

Fingerprint

Cite this