metadata
dataset_info:
features:
- name: id
dtype: string
- name: transcription_channel_1
list:
- name: normalized
dtype: string
- name: original
dtype: string
- name: timestamp
sequence: string
- name: transcription_channel_2
list:
- name: normalized
dtype: string
- name: original
dtype: string
- name: timestamp
sequence: string
- name: audio_codes
sequence:
sequence: int32
splits:
- name: train
num_bytes: 1694654968
num_examples: 11699
download_size: 805330085
dataset_size: 1694654968
configs:
- config_name: default
data_files:
- split: train
path: data/train-*