metadata
dataset_info:
features:
- name: ref_text
sequence: string
- name: ref_spk
sequence: string
- name: hyp_text
sequence: string
- name: hyp_spk
sequence: string
- name: utterance_id
sequence: string
- name: ref_spk_degraded
sequence: string
- name: hyp_spk_oracle
sequence: string
splits:
- name: train
num_bytes: 12960316
num_examples: 136
- name: validation
num_bytes: 1541659
num_examples: 18
- name: test
num_bytes: 1463324
num_examples: 16
download_size: 5312905
dataset_size: 15965299
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*