The dataset could not be loaded because the splits use different data file formats, which is not supported. Read more about the splits configuration. Click for more details.

Couldn't infer the same data file format for all splits. Got {'train': ('parquet', {}), 'test': ('audiofolder', {}), 'validation': ('audiofolder', {})}

Error code:   FileFormatMismatchBetweenSplitsError

Need help to make the dataset viewer work? Make sure to review how to configure the dataset viewer, and open a discussion for direct support.

Nepali ASR Audio Dataset (Noisy)

Dataset Description

This dataset contains Nepali speech audio recordings with corresponding text transcriptions, intended for training and evaluating Automatic Speech Recognition (ASR) models for the Nepali language.

Dataset Summary

Split	Samples
Train	16,809
Test	990
Validation	1,977
Total	19,776

Audio Properties

Format: WAV
Sampling rate: 16,000 Hz
Channels: Mono (1)
Language: Nepali (ne)

Dataset Structure

Data Fields

Field	Type	Description
`file_name`	string	Relative path to the audio WAV file
`transcript`	string	Nepali text transcription (Devanagari script)
`speaker`	string	Anonymized speaker identifier
`duration`	float	Audio duration in seconds
`sampling_rate`	int	Audio sampling rate (16000 Hz)
`channels`	int	Number of audio channels (1 = mono)
`noise_type`	string	Type of noise applied (`none` for clean audio)
`snr_db`	float	Signal-to-noise ratio in dB (-1 for clean)

Data Splits

Each split folder contains:

metadata.jsonl — One JSON object per line with all metadata fields
data.parquet — The same metadata in Apache Parquet format
*.wav — Individual audio files

Usage

Loading with 🤗 Datasets

from datasets import load_dataset

# Load from HuggingFace Hub
dataset = load_dataset("gam30/nepali_noisy_audio")

# Access a sample
sample = dataset["train"][0]
print(sample["transcript"])
# Play audio (in a notebook)
# from IPython.display import Audio
# Audio(sample["audio"]["array"], rate=sample["audio"]["sampling_rate"])

Loading from local directory

from datasets import load_dataset

dataset = load_dataset("audiofolder", data_dir="./huggingface_ready")
print(dataset)