metadata
dataset_info:
features:
- name: file_name
dtype: string
- name: audio_ground_truth
dtype:
audio:
sampling_rate: 44100
- name: audio_xcodec2_recon
dtype:
audio:
sampling_rate: 44100
- name: audio_mel_recon_bigvgan
dtype:
audio:
sampling_rate: 44100
- name: audio_diffusion_generated
dtype:
audio:
sampling_rate: 44100
- name: sampling_rate
dtype: int32
splits:
- name: train
num_bytes: 348564104
num_examples: 267
download_size: 221723222
dataset_size: 348564104
configs:
- config_name: default
data_files:
- split: train
path: data/train-*