metadata
dataset_info:
features:
- name: audio
dtype: audio
- name: transcription
dtype: string
- name: Speaker ID
dtype: string
- name: retext
dtype: string
- name: whisper-med
dtype: string
- name: whisper-large
dtype: string
- name: zipformer-2025-vi
dtype: string
- name: zipformer-multi-stream
dtype: string
splits:
- name: train
num_bytes: 60441568.16
num_examples: 2858
- name: validation
num_bytes: 59668495.84
num_examples: 2912
- name: test
num_bytes: 71852907.705
num_examples: 3437
download_size: 186987674
dataset_size: 191962971.70499998
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*