CoolFace
Datasetpublic

aimeri/ticuna-spanish-portuguese

Ticuna (tca) – Spanish – Portuguese Corpus First text corpus for Ticuna (ISO 639-3 tca), a tonal language isolate of the Brazil/Colombia/Peru tri-border. Configs | Config | Rows | | parallel | train 43,248 / validation 596 / test 3,238 | | monolingual | train 46,545 / validation 298 / test 1,613 | | lexicon | train 10,419 / validation 568 / test 539 | | instructions | train 52,836 | | backtranslation | train 33,944 | The short version of what matters… See the full description on the dataset page: https://huggingface.co/datasets/aimeri/ticuna-spanish-portuguese.

sourceHugging Faceapache-2.0updated 29d agoView on Hugging Face
0likes105downloads
Dataset Card

Ticuna (tca) – Spanish – Portuguese Corpus

First text corpus for Ticuna (ISO 639-3 tca), a tonal language isolate of the Brazil/Colombia/Peru tri-border.

Configs

| Config | Rows | | parallel | train 43,248 / validation 596 / test 3,238 | | monolingual | train 46,545 / validation 298 / test 1,613 | | lexicon | train 10,419 / validation 568 / test 539 | | instructions | train 52,836 | | backtranslation | train 33,944 |

The short version of what matters

  • —Orthography is tagged: orthography_profile per row (tca-PE-2011, tca-BR-sbb, tca-PE-school, …). Do not mix blindly; condition on it.
  • —Splits are whole Bible books (test: MRK PHP RUT JON; dev/validation: 1TH TIT JOL) — the same verse in two editions always shares a split;
  • —Filter for clean training: quality_flag == "ok"; for monolingual also default_train == true (gates unref'd Bible-crawl text and damaged extractions).
  • —Lexicon tone_notation carries the Anderson dictionary's superscript tone digits — the only machine-readable tone data for this language.
  • —Normalization was mechanical only (NFC, exotic whitespace, word-internal apostrophe → U+02BC) and is logged; raw text lives in text_raw columns of the pipeline intermediates.
aimeri/ticuna-spanish-portuguese · CoolFace