Kukedlc's picture
Upload dataset
24c36be verified
|
raw
history blame
714 Bytes
metadata
dataset_info:
  features:
    - name: file_name
      dtype: string
    - name: text
      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
  splits:
    - name: train
      num_bytes: 1642419
      num_examples: 4000
  download_size: 1041995
  dataset_size: 1642419
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*