CoolFace
Datasetpublic

abhiram4572/VeriSpeak

VeriSpeak VeriSpeak is a spoken-statement factual-verification benchmark. Each example is a short synthesized speech clip of a single declarative sentence about a public figure, labeled correct or incorrect depending on whether the spoken statement is factually true. The task: given the audio (and optionally its transcript), decide whether the claim it makes is accurate. It targets speech-native fact-checking / hallucination detection. Dataset at a glance… See the full description on the dataset page: https://huggingface.co/datasets/abhiram4572/VeriSpeak.

sourceHugging Facemitupdated 27d agoView on Hugging Face
1likes21downloads
Dataset Card

VeriSpeak

VeriSpeak is a spoken-statement factual-verification benchmark. Each example is a short synthesized speech clip of a single declarative sentence about a public figure, labeled correct or incorrect depending on whether the spoken statement is factually true.

The task: given the audio (and optionally its transcript), decide whether the claim it makes is accurate. It targets speech-native fact-checking / hallucination detection.

Dataset at a glance

Examples3,879
Splittest only (evaluation benchmark)
Label balance1,973 correct / 1,906 incorrect
Categorieslocation 2,226 · year 1,451 · relation 202
Unique subjects659 (each a distinct public figure)
Audiomono WAV, 22,050 Hz, ~10 s average, ~10.9 h total
LanguageEnglish

Data fields

fieldtypedescription
audioAudio(sampling_rate=22050)the spoken clip; decodes to {array, sampling_rate}
transcriptstringthe exact sentence that was spoken
labelstring"correct" or "incorrect" — is the statement factually accurate
categorystringkind of fact asserted: "location", "year", or "relation"
namestringthe public figure the statement is primarily about
keystringstable id for the subject entity; shared by every clip about the same person
file_namestringoriginal filename, formatted {key}_{category}_{label}_{index}.wav

incorrect clips are minimal perturbations of a true statement — swapping the year, the place, or the related entity — so the surface form stays close to the correct version and the label hinges on the fact, not the phrasing (e.g. "…served as Prime Minister of Belgium from 2008 to 2009." → "…from 2017 to 2011.").

Usage

python
from datasets import load_dataset

ds = load_dataset("abhiram4572/VeriSpeak", split="test")
ex = ds[0]

ex["audio"]        # {'array': array([...]), 'sampling_rate': 22050}
ex["transcript"]   # 'Herman Van Rompuy served as Prime Minister of Belgium from 2008 to 2009.'
ex["category"]     # 'year'
ex["label"]        # 'correct'

Decoding the audio column needs an audio backend: with datasets >= 4.0 install torchcodec, or pin pip install "datasets<4.0" soundfile. The raw bytes are embedded in the Parquet files, so no separate audio download is required.

Filter by category or label

python
years = ds.filter(lambda r: r["category"] == "year")
wrong = ds.filter(lambda r: r["label"] == "incorrect")

# turn `label` into an integer class column if a model needs it
ds = ds.class_encode_column("label")

Source

VeriSpeak is derived from the publicly available KVQA knowledge base, whose public-figure information is sourced from Wikidata. It does not include private or user-provided personal data.

Ethics & intended use

The benchmark contains synthetic false claims solely for controlled evaluation. Each claim is paired with a veracity label and is clearly identified as benchmark-generated in this documentation and in the dataset metadata. VeriSpeak is intended for evaluating speech-based fact-verification systems and should not be treated as a source of factual claims about individuals.

License

MIT.