vamoruso/Corpus_biscegliese

📚 Corpus Biscegliese — Dialect Dataset (JSONL)

The Corpus Biscegliese is a JSONL dataset designed for training, evaluating, and studying language models specialized in the Biscegliese dialect, a local linguistic variety spoken in Bisceglie (Apulia, Southern Italy).

This dataset was used to train the model:
👉 Biscegliese-Qwen2.5-GGUF
https://huggingface.co/vamoruso/biscegliese-qwen2.5-gguf

---

📌 Dataset Structure

The dataset is provided as a JSON Lines (JSONL) file.
Each line contains an independent example in the following format:

{"italiano": "<text in Italian>", "biscegliese": "<translation in Biscegliese dialect>"}