asr-task-hindi / README.md
hamees's picture
Upload dataset
f9faa4c verified
metadata
dataset_info:
  features:
    - name: audio
      dtype:
        audio:
          sampling_rate: 16000
    - name: text
      dtype: string
  splits:
    - name: train
      num_bytes: 229160695
      num_examples: 8000
  download_size: 214778507
  dataset_size: 229160695
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*