real_fake_audio

Dataset Description

This is an audio dataset for deepfake detection tasks.

Dataset Summary

• Task: deepfake
• Audio Format: WAV/FLAC
• Sample Rate: 16kHz/22.05kHz
• License: [Specify your license]

Usage

from datasets import load_dataset

dataset = load_dataset("CosminMihai02/real_fake_audio")

# Access audio
audio_sample = dataset['train'][0]['audio']
print(f"Sample rate: {audio_sample['sampling_rate']}")
print(f"Audio shape: {audio_sample['array'].shape}")

Data Fields

• audio: Audio file
• filename: Original filename
• duration: Duration in seconds
• [Add other fields based on your dataset]

Citation

@dataset{CosminMihai02_real_fake_audio,
  title={real_fake_audio},
  author={Your Name},
  year={2024}
}