metadata
dataset_info:
features:
- name: id
dtype: string
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: sentence
dtype: string
- name: target_word
dtype: string
- name: target_word_index
dtype: int64
- name: alignments_words
dtype: string
- name: alignments_phonemes
dtype: string
- name: ipa_phoneme_sequence
dtype: string
- name: arpa_phoneme_sequence
dtype: string
- name: gen_start_time
dtype: float64
- name: gen_end_time
dtype: float64
- name: target_phonemes_ipa
dtype: string
- name: target_phonemes_arpa
dtype: string
- name: description
dtype: string
- name: llasa_sequence
dtype: string
splits:
- name: train
num_bytes: 2555974
num_examples: 25
download_size: 2374023
dataset_size: 2555974
configs:
- config_name: default
data_files:
- split: train
path: data/train-*