metadata
dataset_info:
features:
- name: utterance_id
dtype: string
- name: audio_bytes
dtype: binary
- name: transcription
dtype: string
- name: raw_text
dtype: string
- name: speaker_id
dtype: string
- name: duration
dtype: float32
- name: language
dtype: string
- name: gender
dtype: string
splits:
- name: train
num_bytes: 44843692
num_examples: 2
- name: test
num_bytes: 55882847
num_examples: 1
download_size: 100564763
dataset_size: 100726539
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*