opus-mt-en-rhg

English → Rohingya (opus-mt finetune)

Translates English into Rohingya. Finetuned by Linguardia from Helsinki-NLP/opus-mt-en-mul on the English↔Rohingya sentence pairs in Tatoeba.

Built to generate course material for a language that no major machine translator supports.

Usage

from transformers import AutoTokenizer, AutoModelForSeq2SeqLM

tok = AutoTokenizer.from_pretrained("linguardia/opus-mt-en-rhg")
model = AutoModelForSeq2SeqLM.from_pretrained("linguardia/opus-mt-en-rhg")

batch = tok([">>rhg<< I wake up early."], return_tensors="pt")
out = model.generate(**batch, num_beams=4, max_length=64)
print(tok.batch_decode(out, skip_special_tokens=True))

The target language is selected by the >>rhg<< prefix on the source text. It was added as a new token rather than replacing one of the base model's existing languages, so every language the base already handled still works.

Training

Evaluation

Held-out Tatoeba (300 pairs never seen in training):

Translating the same 2,000-sentence course corpus:

Training data and attribution

Sentence pairs come from Tatoeba, released under CC-BY 2.0 FR, with some sentences under CC0 1.0. Tatoeba's sentences are written by volunteers, and this model is a derivative of their work: if you use it, credit Tatoeba and its contributors.

Pairs were deduplicated on exact (English, Rohingya) text. No machine translation was used to create the training data at any point.

Licence chain

Limitations

• Trained on Tatoeba, which is short everyday sentences. Expect it to be weakest on long, technical or literary text.
• Neural MT translates each sentence independently, so a term may be rendered differently across a corpus. Check consistency if that matters to you.
• Output should be reviewed by a speaker before being taught to learners.