Datasets:
metadata
dataset_info:
features:
- name: speaker_id
dtype: string
- name: transcription_id
dtype: int64
- name: text
dtype: string
- name: audio
dtype:
audio:
sampling_rate: 44100
splits:
- name: train
num_bytes: 12163543668.45736
num_examples: 18863
download_size: 10460673849
dataset_size: 12163543668.45736
configs:
- config_name: default
data_files:
- split: train
path: data/train-*