metadata
dataset_info:
features:
- name: language
dtype: string
- name: speaker_id
dtype: string
- name: prompt_id
dtype: string
- name: prompt
dtype: string
- name: segment_id
dtype: string
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: duration
dtype: float64
- name: raw_text
dtype: string
- name: iso_639_3
dtype: string
- name: glottocode
dtype: string
- name: iso_15924
dtype: string
splits:
- name: train
num_bytes: 11909146889
num_examples: 12317
- name: test
num_bytes: 1323668285
num_examples: 1369
download_size: 13229977056
dataset_size: 13232815174
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*