Twi-English Parallel Dataset

This dataset contains a large-scale parallel corpus of Twi-English sentence pairs, featuring synthetically generated Twi sentences with corresponding English paraphrases. The dataset is designed to support machine translation, cross-lingual understanding, and other NLP tasks involving the Twi language (a dialect of Akan spoken in Ghana). Total Size: 47,924,398 parallel sentence pairs