metadata
dataset_info:
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: transcription
dtype: string
- name: translation
dtype: string
- name: file
dtype: string
splits:
- name: test
num_bytes: 1028671749
num_examples: 14
download_size: 1008169799
dataset_size: 1028671749
configs:
- config_name: default
data_files:
- split: test
path: data/test-*
Derived from: ky552/ML2021_ASR_ST Segments from the same lecture are concatenated together.