Datasets:

Multilinguality: multilingual
Size Categories: 100K<n<1M
Licenses: cc-by-4.0
Language Creators: found
Annotations Creators: expert-generated
Source Datasets: extended|wikipedia

Dataset Card for "bbaw_egyptian"

Dataset Summary

This dataset comprises parallel sentences of hieroglyphic encodings, transcription and translation as used in the paper Multi-Task Modeling of Phonographic Languages: Translating Middle Egyptian Hieroglyph. The data triples are extracted from the digital corpus of Egyptian texts compiled by the project "Strukturen und Transformationen des Wortschatzes der ägyptischen Sprache".

Supported Tasks and Leaderboards

More Information Needed

Languages

The dataset consists of parallel triples of

  • hieroglyphs: [Encoding of the hieroglyphs with the Gardiner's sign list
  • transcription: Transliteration of the above mentioned hieroglyphs with a transliteration scheme
  • translation: Translation in mostly German language (with some English mixed in)

Dataset Structure

The dataset is not divided into 'train', 'dev' and 'test' splits as it was not built for competitive purposes and we encourage all scientists to use individual partitioning schemes to suit their needs (due to the low resource setting it might be advisable to use cross validation anyway). The only available split 'all' therefore comprises the full 100,708 translation triples, 35,503 of which possess hieroglyphic encodings (the remaining 65,205 triples have empty hieroglyph entries).

Data Instances

An example of a data triple looks the following way:

{
    "transcription": "n rḏi̯(.w) gꜣ =j r dbḥ.t m pr-ḥḏ",
    "translation": "I was not let to suffer lack in the treasury with respect to what was needed;",
    "hieroglyphs": "D35 D21 -D37 G1&W11 -V32B A1 D21 D46 -D58 *V28 -F18 *X1 -A2 G17 [? *O2 *?]"
}

Important: Only about a third of the instance actually cover hieroglyphic encodings (the rest is the empty string "") as the leftover encodings have not yet been incorporated into the BBAW's project database.

Data Fields

plain_text

  • transcription: a string feature.
  • translation: a string feature.
  • hieroglyphs: a string feature.

Data Splits

name all
plain_text 100708

Dataset Creation

Curation Rationale

More Information Needed

Source Data

Initial Data Collection and Normalization

The data source comes from the project "Strukturen und Transformationen des Wortschatzes der ägyptischen Sprache" which is compiling an extensively annotated digital corpus of Egyptian texts. Their publication comprises an excerpt of the internal database's contents.

Who are the source language producers?

[More Information Needed]

Annotations

Annotation process

The corpus has not been preprocessed as we encourage every scientist to prepare the corpus to their desired needs. This means, that all textcritic symbols are still included in the transliteration and translation. This concerns the following annotations:

  • (): defective
  • []: lost
  • {}: surplus
  • 〈〉: omitted
  • ⸢⸣: damaged
  • ⸮?: unclear
  • {{}}: erasure
  • (()): above
  • [[]]: overstrike
  • 〈〈〉〉: haplography

Their exists a similar sign list for the annotation of the hieroglyphic encoding. If you wish access to this list, please get in contact with the author.

Who are the annotators?

[More Information Needed]

Personal and Sensitive Information

More Information Needed

Considerations for Using the Data

Social Impact of Dataset

More Information Needed

Discussion of Biases

More Information Needed

Other Known Limitations

More Information Needed

Additional Information

Dataset Curators

More Information Needed

Licensing Information

More Information Needed

Citation Information

Source corpus:

@misc{OPUS4-2919,
  title     = {Teilauszug der Datenbank des Vorhabens "Strukturen und Transformationen des Wortschatzes der {\"a}gyptischen Sprache" vom Januar 2018},
  institution = {Akademienvorhaben Strukturen und Transformationen des Wortschatzes der {\"a}gyptischen Sprache. Text- und Wissenskultur im alten {\"A}gypten},
  type      = {other},
  year        = {2018},
}

Translation paper:

@article{wiesenbach19,
  title = {Multi-Task Modeling of Phonographic Languages: Translating Middle Egyptian Hieroglyphs},
  author = {Wiesenbach, Philipp and Riezler, Stefan},
  journal = {Proceedings of the International Workshop on Spoken Language Translation},
  journal-abbrev = {IWSLT},
  year = {2019},
  url = {https://www.cl.uni-heidelberg.de/statnlpgroup/publications/IWSLT2019_v2.pdf}
}

Contributions

Thanks to @phiwi for adding this dataset.

Models trained or fine-tuned on bbaw_egyptian

None yet