inference / README.md
Cafet's picture
Upload dataset
d4f27c5 verified
metadata
dataset_info:
  features:
    - name: audio
      dtype:
        audio:
          sampling_rate: 16000
    - name: transcription
      dtype: string
    - name: input_features
      sequence:
        sequence: float32
    - name: input_length
      dtype: int64
    - name: reference
      dtype: string
    - name: prediction_whisper_tiny
      dtype: string
    - name: prediction_w2v_test
      dtype: string
    - name: prediction_whisper_large
      dtype: string
    - name: prediction_whisper_small
      dtype: string
  splits:
    - name: validation
      num_bytes: 106051608
      num_examples: 194
  download_size: 104264192
  dataset_size: 106051608
configs:
  - config_name: default
    data_files:
      - split: validation
        path: data/validation-*