metadata
dataset_info:
features:
- name: audio
struct:
- name: array
sequence: float64
- name: path
dtype: string
- name: sample_rate
dtype: int64
- name: text
dtype: string
- name: scriptId
dtype: int64
- name: fileNm
dtype: string
- name: recrdTime
dtype: float64
- name: recrdQuality
dtype: int64
- name: recrdDt
dtype: string
- name: scriptSetNo
dtype: string
- name: recrdEnvrn
dtype: string
- name: colctUnitCode
dtype: string
- name: cityCode
dtype: string
- name: recrdUnit
dtype: string
- name: convrsThema
dtype: string
- name: gender
dtype: string
- name: recorderId
dtype: string
- name: age
dtype: int64
- name: word_timing_info
list:
- name: end_time
dtype: float64
- name: index
dtype: int64
- name: start_time
dtype: float64
- name: word
dtype: string
splits:
- name: train
num_bytes: 9457703418
num_examples: 12401
download_size: 2076254461
dataset_size: 9457703418
configs:
- config_name: default
data_files:
- split: train
path: data/train-*