Datasetpublic

lopozz/accenti_italiani

🤌 Accenti Italiani Accenti Italiani is a dataset designed to evaluate Automatic Speech Recognition (ASR) models under challenging real-world conditions.It was created to test model robustness on strong regional Italian accents together with noisy audio, making the task deliberately difficult. Overview Format: WAV Sampling rate: 16000 Hz How to load from datasets import load_dataset from IPython.display import Audio, display ds =… See the full description on the dataset page: https://huggingface.co/datasets/lopozz/accenti_italiani.

sourceHugging Faceupdated 5mo agoView on Hugging Face

0likes10downloads

Dataset Card

🤌 Accenti Italiani

Accenti Italiani is a dataset designed to evaluate Automatic Speech Recognition (ASR) models under challenging real-world conditions. It was created to test model robustness on strong regional Italian accents together with noisy audio, making the task deliberately difficult.

Overview

—Format: WAV
—Sampling rate: 16000 Hz

How to load

python

from datasets import load_dataset
from IPython.display import Audio, display

ds = load_dataset("lopozz/accenti_italiani", split="train")

example = ds[0]
display(Audio(example["audio"]["array"], rate=example["audio"]["sampling_rate"]))

print(example["text"])
print(example["source"])

Citation

Please cite this dataset as follows:

@misc{accenti_italiani,
  author = {Lorenzo, Pozzi},
  title = {AccentiItaliani: An Authentically Italian Dataset for ASR},
  year = {2026},
  publisher = {Lorenzo Pozzi},
  url = {https://huggingface.co/datasets/lopozz/accenti_italiani}
}