Merged_Luo_Dataset / README.md
KYAGABA's picture
Upload dataset
b4807cc verified
metadata
dataset_info:
  features:
    - name: audio
      dtype:
        audio:
          sampling_rate: 16000
    - name: transcription
      dtype: string
  splits:
    - name: train
      num_bytes: 2332757832.46
      num_examples: 6435
    - name: test
      num_bytes: 224637696.579
      num_examples: 2797
    - name: validation
      num_bytes: 85968232.574
      num_examples: 3162
  download_size: 2553050674
  dataset_size: 2643363761.613
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*
      - split: test
        path: data/test-*
      - split: validation
        path: data/validation-*