metadata
dataset_info:
features:
- name: id
dtype: string
- name: path
dtype:
audio:
sampling_rate: 16000
- name: content_enjoyment
dtype: float64
- name: content_usefulness
dtype: float64
- name: production_complexity
dtype: float64
- name: production_quality
dtype: float64
- name: sample_rate
dtype: int64
- name: n_channels
dtype: int64
- name: pesq_hyp
dtype: float64
- name: stoi_hyp
dtype: float64
- name: si_sdr_hyp
dtype: float64
- name: num_channels
dtype: float64
- name: duration
dtype: float64
- name: num_speakers
dtype: int64
splits:
- name: train
num_bytes: 832765512
num_examples: 5000
download_size: 846888361
dataset_size: 832765512
configs:
- config_name: default
data_files:
- split: train
path: data/train-*