WenetSpeech_Wu_1k / README.md
yuekai's picture
Upload dataset (part 00004-of-00005)
940605e verified
|
Raw History Blame Contribute Delete
942 Bytes
metadata
dataset_info:
  features:
    - name: utt_id
      dtype: string
    - name: audio
      dtype:
        audio:
          sampling_rate: 16000
    - name: duration
      dtype: float64
    - name: text
      dtype: string
    - name: translation
      dtype: string
    - name: confidence
      dtype: float64
    - name: DNSMOS
      dtype: float64
    - name: SNR
      dtype: float64
    - name: speech_rate
      dtype: float64
    - name: loudness
      dtype: float64
    - name: f0_mean
      dtype: float64
    - name: f0_std
      dtype: float64
    - name: energy_mean
      dtype: float64
    - name: energy_std
      dtype: float64
    - name: domain
      dtype: string
    - name: sub_dialect
      dtype: string
    - name: is_multispeaker
      dtype: string
  splits:
    - name: train
      num_bytes: 116197024158.722
      num_examples: 478639
  download_size: 106076094285
  dataset_size: 116197024158.722
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*