metadata
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
dataset_info:
features:
- name: flac
dtype: audio
- name: txt
dtype: string
- name: json
struct:
- name: uid
dtype: string
- name: source_uid
dtype: string
- name: chapter_uid
dtype: string
- name: book_title
dtype: string
- name: chapter_index
dtype: int64
- name: chapter_title
dtype: string
- name: audio_sample_rate
dtype: int64
- name: speaker
dtype: string
- name: clip_index
dtype: int64
- name: clip_duration_sec
dtype: float64
- name: diarization_backend
dtype: string
- name: diarization_num_speakers
dtype: int64
- name: transcription_backend_name
dtype: string
- name: transcription_language
dtype: string
- name: char_count
dtype: int64
- name: cps
dtype: float64
- name: quality_device
dtype: string
- name: speaker_global
dtype: string
- name: __key__
dtype: string
splits:
- name: train
num_bytes: 3238183292688
num_examples: 10758183
download_size: 3238322530572
dataset_size: 3238183292688