metadata
dataset_info:
features:
- name: audio
dtype: audio
- name: transcript
dtype: string
- name: greennode_asr
dtype: string
- name: latency
dtype: string
- name: WER
dtype: 'null'
- name: CER
dtype: 'null'
- name: lower_WER
dtype: 'null'
- name: lower_CER
dtype: 'null'
- name: source
dtype: string
- name: audio_id
dtype: string
splits:
- name: train
num_bytes: 90388175
num_examples: 300
download_size: 88176735
dataset_size: 90388175
configs:
- config_name: default
data_files:
- split: train
path: data/train-*