metadata
dataset_info:
features:
- name: file_name
dtype: string
- name: text
dtype: string
- name: speaker_name
dtype: string
- name: transcription_normalised
dtype: string
- name: utterance_pitch_mean
dtype: float32
- name: utterance_pitch_std
dtype: float32
- name: snr
dtype: float64
- name: c50
dtype: float64
- name: speaking_rate
dtype: float64
- name: phonemes
dtype: string
- name: stoi
dtype: float64
- name: si-sdr
dtype: float64
- name: pesq
dtype: float64
- name: gender
dtype: string
splits:
- name: train
num_bytes: 197340504
num_examples: 435332
- name: test
num_bytes: 49345153
num_examples: 108835
download_size: 164872450
dataset_size: 246685657
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*