metadata
dataset_info:
features:
- name: text
dtype: string
- name: nano_layer_1
list: int64
- name: nano_layer_2
list: int64
- name: nano_layer_3
list: int64
- name: nano_layer_4
list: int64
- name: encoded_len
dtype: int64
- name: speaker
dtype: string
- name: wavlm_embedding
list: float64
- name: grouped_sp_emb
list: float64
- name: cosine_sim
dtype: float64
splits:
- name: train
num_bytes: 102348484
num_examples: 20153
download_size: 41569621
dataset_size: 102348484
configs:
- config_name: default
data_files:
- split: train
path: data/train-*