This dataset provides a large collection of English-French translation pairs. It was downloaded from the OPUS collection on 2025-02-26.
These datasets cover basic French, news, presentations (like TED talks and academic conferences), and educational videos. This helps avoid the bias towards legal, administrative, and bureaucratic content found in datasets like Europarl and ECB.
Normalization: Bullet points, hyphenated words… See the full description on the dataset page:
https://huggingface.co/datasets/agentlans/en-fr.
