---
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: 106051608.0
    num_examples: 194
  download_size: 104264192
  dataset_size: 106051608.0
configs:
- config_name: default
  data_files:
  - split: validation
    path: data/validation-*
---