Datasetpublic

drewoodward/spanglish-sentences

Spanglish Sentences A dataset of 10,576 Spanish–English code-switched ("Spanglish") sentences paired with English translations, intended for training and evaluating code-switch translation models. Data format Each line of spanglish_sentences.jsonl is a JSON object with two fields: field description sentence A Spanglish utterance (mixed Spanish / English, or monolingual in either language). english_translation The English translation. When the source… See the full description on the dataset page: https://huggingface.co/datasets/drewoodward/spanglish-sentences.

sourceHugging Facecc-by-4.0updated 5mo agoView on Hugging Face

0likes51downloads

Dataset Card

Spanglish Sentences

A dataset of 10,576 Spanish–English code-switched ("Spanglish") sentences paired with English translations, intended for training and evaluating code-switch translation models.

Data format

Each line of spanglish_sentences.jsonl is a JSON object with two fields:

field	description
`sentence`	A Spanglish utterance (mixed Spanish / English, or monolingual in either language).
`english_translation`	The English translation. When the source is already English, it is reproduced unchanged.

Example:

json

{"sentence": "él podía escoger o una inyección o unas pastillas", "english_translation": "He could either pick an injection or some pills."}
{"sentence": "yeah como un asesino porque ya ellos tenían su comunidad ahí", "english_translation": "Yeah like a killer because they already had their community there."}

Provenance

Sentences and their English translations were generated by a large language model. This is a synthetic dataset; none of the content corresponds to real speakers or recordings.

Intended use

Training and evaluating code-switch translation systems (Spanglish → English).

Limitations

—Synthetic: linguistic patterns may not faithfully reflect natural Spanglish usage in any specific community (Miami, Caribbean, Mexican-American, Chicano, etc.). Evaluate against a human-produced test set before drawing conclusions about real-world performance.
—Translation quality is LLM-generated and has not been human-verified. Expect noise, including cases where the "translation" simply copies the source.
—Punctuation, capitalization, and orthography are inconsistent (some sentences lack punctuation entirely, some mix casing).
—Many lines are short fillers ("yeah", "you know") where source and translation are identical — filter these out if your task requires non-trivial translation pairs.

License

Released under the Creative Commons Attribution 4.0 license (CC BY 4.0).