metadata
dataset_info:
config_name: zh-TW
features:
- name: client_id
dtype: string
- name: path
dtype: string
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: sentence
dtype: string
- name: up_votes
dtype: int64
- name: down_votes
dtype: int64
- name: age
dtype: string
- name: gender
dtype: string
- name: accent
dtype: string
- name: locale
dtype: string
- name: segment
dtype: string
- name: variant
dtype: string
- name: whisper_transcript
sequence: int64
splits:
- name: train
num_bytes: 160498655.447
num_examples: 6799
- name: validation
num_bytes: 122880246.375
num_examples: 4825
- name: test
num_bytes: 142152848.375
num_examples: 4825
download_size: 398340449
dataset_size: 425531750.197
configs:
- config_name: zh-TW
data_files:
- split: train
path: zh-TW/train-*
- split: validation
path: zh-TW/validation-*
- split: test
path: zh-TW/test-*